Make AI Videos in Kling AI | Generate Text-to-Video and Image-to-Video

1. INTRODUCTION: Discovered the power of Kling AI. The leading video generation tool that transforms simple text and images into professional and captivating videos with ease. Hi, everyone. My name is Faizan Amjad. And in this course, you will learn to generate videos from scratch using stunning features of King AI like text to video and image to video generation. I'll guide you through every step from generating hyperrealistic images to seamlessly integrating them into videos. By the end of this course, you will be equipped with skill to produce stunning video content effortlessly. So what are you waiting for? Join now and unlock your creative potential with Kling AI. 2. How to Get Access Kling AI: Welcome. So to get access to cling AI, search for cling AI, and you have to open this link cling ai.com. Once you click the link, it will take you to this page. This is the interface of cling.com. Here, you can sign in for free, and you will get 66 credit on a daily basis, and the credit will reset on a daily basis as well. Now I'm going to sign in for free. If you have an account, you can add your email and password. But if you don't have an account, sign up for free. Click here. Now click on this next button. Go to your mail and you will get this verification number. Copy and paste this number right here and now click on this sign in button. And now, as you can see, guys, we have 66 credit. Now I am going to introduce you to the interface of cling AI. Here, as you can see, guys, if you want to generate AI images, then you can click here. And if you want to turn these images or any image into videos, then you can click here. You can also generate your AI videos as well, and there will be also a video editor in the feature soon, so make sure you stay tuned. Right here, as you can see guy, these are some example created with ling AI. Now, if you scroll down as you can see guy, these are some AI images and videos created with ling AI. So the user of this website created these examples. So you can get the inspiration from these images and video as well. You can click any image or video, and here as you can see guy, this is the prompt of this image. This is the image ratio, and this is a reference image, which is none. So this user created this image using all in the prompt with no reference image. So this is the interface of ling AI. 3. New Interface of Kling AI: Welcome back everyone. So the link AI change their interface. If you learn ling AI from the previous interface, don't worry, they just polish the interface. They did not completely change it. So if you now go to the Kling AI website, this is a new interface. Now you can hop your mouse to this creative studio, and as you can see in AI Tools, we can see we have video generation, image generation. We even have a new feature called Sound Generation. You can even generate sound from now on. We have effec and if you click here Metols, it will show you all the tools that are available on this website. Now we can simply click here onto this Create button. We can close this tab for now. And now, as you can see, we have this interface. Now we have this homepage. You can explore. We have the asset. If you go to the asset, this is all the videos that I have created using Cling AI, and we have also the images as well. You can go to the effects to create some effects. You can go to the image. We have a new model for the image called colors 2.0. We even have a feature called image editing. We have a new cling 2.0 model. If you go to the videos, as you can see, this is a brand new interface for video generation. This is the text to video. This is image to video. You can even switch the model from here. Like in the previous interface we did. Now this section is completely changed. Now, as you can see, this section is polished. Now we can click here. We can view the images only. We can click here. We can switch to videos, and now it will only show you all the videos that we have created using Clean AI. Now you can select your video from the small thumbnails. You can even show the audio. So currently, I did not generate any audio. That is why we don't have any audio shows here. Right here, as you can see, we have effects. If you click here, you can go to the homepage you can explore or you can even go to your assets. Now right here, we can extend the video. Right here, we have the lip sync. We have AI, which will try on sound generation, we have Video generation. Currently, we are here in the Video generation. That is why we have this green logo right here. If you go to the image generation, now the image generation logo will turn green. So basically, this is a new polish interface of ling AI. 4. How to Generate Image in New Interface: Welcome back, everyone. So the Kling AI recently changed the interface. They just polish the interface. So if you want to generate images in the new interface, simply go to image, and as you can see, we have this interface. Now we have also the videos that are shown here. So if you want to use only the images, you can click here and you can switch to images. And now it will show you only the images that you generated using Kling AI. Now we even have a new model for image generation called colors 2.0. Color 2.0 has an in feature called restyle. So what the restyle does, it will take your image as a reference, and if you type in any style, it will change your image according to that style. Now, right here, we can change the standard mode to high resolution. You can change the output. You can change your resolution aspect ratio from here. Now if you switch to subject, as you can see, it automatically switch to the available models, which is 1.5 colors. If you switch to pass, it will also switch to 1.5 because 2.0 colors model does not spot subject phase and entire image reference. If you switch to entire image reference, the model again switch automatically to colors 1.0 because the entire image reference is only available for colors 1.0. If you switch to restyle, then it will automatically switch to colors 2.0. Now you can type in any prompt and you can generate your image. So this is how you can generate images in the new interface of cling AI. 5. How to Generate Videos in New Interface: Welcome back, everyone. So how do we generate videos in the new interface of Kling AI? So Kling AI recently changed the interface. They just polish the interface for video generation, go to videos. And as you can see, we even have a new model for Kling called Kling 2.0 Master. Now you can go to the text to video, and you can also select the preset. You can select your lens, show type, lighting, frame, atmosphere, et cetera. Now, if you want to use the motion brush feature, you can go to the image to video, and as you can see, we don't have any motion brush feature. So to access to Motion brush feature, you have to click here. You have to change your model to cling 1.5. As you can see, cling 1.5, standard mode does not support the motion brdge. So you have to switch your model to professional, which is for VIPs. Now you can access to Motion brush feature. You can also access to camera movement. So now we have this polish interface. Have this creativity and relevance feature right here. You can select your outputs, and you can even change your second 5-10. Now, if you switch your model to Kling 1.6, now as soon as we switch our model to Kling 1.6, as you can see, we have this message. Currently unavailable on Kling 1.6, please switch to 1.5. Now you don't have to go to this area and switch your model to 1.5. You can switch your model from here. Now we can confirm it. And as you can see, now we switch our model to 1.5, and now we have the motion bridge and camera moments available for us. Now, again, you can go to the elements, and if you switch your model to Kling 2.0 master, currently, it does not sport elements, but we do have multiple elements, and it is also available for Kling 1.6. So Kling 2.0 master model of Kling only sport text to video and image to video 4720 p resolution only. So basically, this is a new interface of Kling AI. 6. Kling AI Interface: Welcome back. Now we will create some basic images so you can better understand cling AI. Click on these AI images. It will take you to this interface. Right here, you can type any prompt. You can generate any image. You can also reference your image if you want to generate a similar result according to your reference. Hey, you can select your aspect ratio, and here you can select number of generation to your specific prompt. Your result will be displayed to this portion of this website. Cling AI take 0.20 credit to generate one image. We are generating four image, so it will take 0.80 credit. For video, it will take ten credit. You will get 66 credit and it will expire in 24 hours. If you, for example, use 20 credit out of this 66, it will reset to 66 the next day. Now let's type a simple prompt. So I'm going to type a glass of water on top of a table. Now, as you can see, guys, if I lower the generation number, it will also lower my credits. So let's generate for generation, and let's select this aspect ratio. Now, we don't have any reference image, so let's click on this generate. Now keep this in mind, we are using the free account, so it will take some time to generate your images or videos. Now you don't have to wait for your generation, you can type your next prompt as well. Now as you can see, guys, we have these images, and these are stunning images. Now you can type any prompt and you can generate any image that you want to generate. Now, as you can see, guys, if I hold my mouse to an image, we have three options. One is enhanced, which is available for the premium user. Don't worry. I have a premium account, so I'm going to show you everything that this website has. Right here, as you can see, we can use this image as a reference, and right here, we can bring this image to life. So I'm going to select this image. Bring this image to life. And as you can see, guys, it will take ten credit to generate this image. Click on this generate button. Now, as I told you earlier, you don't have to sit here and wait for a couple of minutes or maybe hours to watch your image converting into a video. You can generate next video if you want to. So right here, as you can see, guys, if we go back back to the dashboard, let's close this we click on these AI videos. Now right here, as you can see g, this website is now generating our video. Here we have the first option, which is text to video. So basically, this is similar to the AI images. If I hold my mouse right here, we can go to the AI images. We can type any prompt. So this is similar to the text video. Now we can generate any video if you want to just type a prompt. You can select this prompt. For example, you can select your camera moment, so I'm going to select till and now let's generate this video. As you can see, guys, I can generate more than one video and I don't have to wait for it. So you can generate multiple videos at the same time. Now, I know there are so many features of this website. So don't be panic. This is just an introduction to the website. I will introduce you to every feature of cling AI, and we will create some stunning images and videos using this AI tool. So here we have the result. So it turned my image into this beautiful video. Now we have this video as well. We generate this video through this prompt. Now there is a huge difference between the premium account and the free account of King AI. The first difference is you can't use Kling 1.5 if you are using the free account of Kling AI. The second difference, which is very huge, is it took you longer if you are using the free account to generate your clips than the premium account. So, believe me or not, it took me two days to generate this clip. Sometimes it generates your clip very fast and sometimes it generates your clip very slow. To that point that it failed to generate your result. And once it failed to generate your result, it will return your credit. So as you can see, guys, my credit is more than 66. The free account limit is to have only 66 credit for 24 hours. Why don't I have more than 66 credit? Because I have so many failed generation, so it will return your credits and it plus your credit to your current credits. Your credits will expire in 24 hours, no matter if you use all those credits or not, it will reset to 66 credits. So this is just a basic introduction of Kling AI. I have a premium account, so we'll dive deep into this tool, and we will explore its full potential to generate desired results. So I will see you guys in the next lecture. 7. Generate Stunning Images: Welcome back. Now, first of all, click A and go to AI images. There is two ways to generate your image. First is to write your prompt. The second, you can reference an image. If you want to create the similar result, then you can reference an image to create your result according to that image. Now, as you can see, guys, I have generated so many amazing Images in this tool because I write more than few words to generate my impressive results. Now, if you are using the premium account, you can also enhance your image as well. So for example, if you click here, you can hover your mouse, and we have an option called enhance. If I click on it, it will enhance my image. It will turn that image into a higher definition. Now, as you can see, guys, this is my result. This is the previous result, and this is my new result. You can also hover your mouse right here. If you are using the premium account, then you can also download this result without any watermark. In a free account, you will download any image that you want, but you will have a watermark with that image as well. This is a premium account we don't have any watermark. And if you use a free account, then you will have a watermark like this one. Now if you want to create impressive results, then you have to write more than a few words. So how can you find your words? Go to designer.miicrosoft.com, sign in, click here, and you have to select images. Now we can close this tab. Now what you have to do, you have to click here and you have to write your prompt. So for example, I'm going to write. So this is just a simple prompt closeup of a burger in a hotel. Now there is an option called enhanced prompt. And what it does, if I click on this enhanced prompt, it added few words to make this prompt impressive. Now we can copy this prompt. We can go to the killing AI. Let's paste this prompt, and we can select our resolution. You can select your ratio. So I'm going to select this one. You can select the number of generation that you want to generate. I'm going to select four. And now let's click on this generate. Now, let's go back to Microsoft Designer and we can select any resolution right here. Now we can click on this generate button and it will generate this result. These are all our images that we generated through this prompt. Now we can enhance and we can also reference that image as well if you want to. We can also bring this image to life as well. Now, this is the Microsoft designer result, and I like this one. Then the ling AI. So what we can do, we can just click on this download, go to the ling AI. We can click A to reference that image. Now before we generate the result according to the reference image, there is a slider. We have zero plus, and we have three plus, right here, we have minus and we have three minus. So this is the strength of your reference. If you place your strength to the zero position, it will generate your result according to this exact image. If you place your slider to the plus side, it will just add something to your image. And if you place your slider to the minus side, then it will take something from By, I'm going to show you all the results according to this slider. So first of all, let's try this zero position. And by the way, if you reference any image, then it will take 0.40 credits to reference that image. So without any reference, you can generate four images in 0.80 credits, but we have a reference, so now we have 1.20 credits. Now, click on this generate button to generate the results according to the reference. We don't have to wait until the result is generating. We can take this slider to the plus side and now let's click on this strand button. Now I can take the slider to the three plus size, and now we can generate results. Now take the slider to the minus. Now we have these results. The strength is set to zero, and the results are amazing. Now, if you set your strength plus, it will edit something to your images. As you can see, guys, if you go to the original image with the strength set to zero, don't have anything in our plates, but if you go to the plus strength, as you can see, guys, it just added something into your image. So we have these bread pieces added to our burger. And if we go to the le plus, it would add something more to your image. In the triple minus, it just ignore my hotel room and it place my burger onto a table. And in just one minus, it will just take something from the image. So to explain this strength, if you set your strength of zero, it will just generate the exact same result according to your reference. But if you place your slider to the plus side, it will addit something to your image. If you take your slider to the left side, then it will take something from your image. There is another example that I want to explain to you. So with the prompt, I have generated these results. With the reference image with the strength set to plus sign, as you can see, guys, it just edit something in my image. With the strength set to zero, it does generate the exact same results. And with this strength set to the triple minus, it takes everything from my image and it just generate the vector of the image. Now, when I was revisiting my results, I noticed something. With this strength zero, you can see we have these normal images, but with this strength set to the plus, check out the size of the burger. The burger size is also getting bigger. So this is a normal size. This is one plus size, and this is three plus size. Check out the burger size. The burger size is also getting bigger. Now, if we go to the three minus, the burger size is getting smaller. And this is only one minus result. Now, in my second example, I also notice this as well. With this reference, my goals is slim, but with the strength set to three pluses, as you can see, guys, my gold is getting fatty. With zero strength, we have normal size, and with minus sine, it getting smaller and smaller. And it also remove something from my reference, and it just generate this image. So you have to also keep this in mind when you use reference strength to generate your image. This is how you can generate AI images in cling.com. Goodbye. I will see you in my next lecture. 8. AI Virtual Try-On: Welcome back, everyone. So if you go to the AI images, we have a new feature called AI virtual Tryon. So you can try on any shirt, any pen, you can try on any garments onto yourself. Or if you scroll down, you can use this model as well. And as you can see, guys, these are some examples. So we have two options. First, we have single garments, and then we have multiple garments. So first of all, I am going to select this shirt. Now, if you scroll down, I am going to use this model. And now let's generate. Let's generate one output. And as you can see, guys, it is now generating this garment onto this model. And as you can see, guys, just like that, we have this garment onto this person or onto this model. Now let's try on multiple garments. So first of all, I'm going to upload the top, so I'm going to select this shirt. You don't have to remove the background of the garment. You can use this image as well. Now I can use this black pen. Now if you scroll down, we have another option called custom now I'm going to upload my image right here. Now the image is uploaded successfully. Let's generate this garment onto myself. And as you can see, guys, we have this shirt and paint right here. Now, if you hover your mouth to your image, we have another option called Expand. Now we can click on it and we can expand to a ratio. So I'm going to select this ratio and let's click on Expand image immediately. This way, you can change your vertical images to horizontal you can change your image to any other aspect ratio. And as you can see guide, this is the final output. Now I'm going to show you another example. So as you can see guide, this is the bottom and this is the top. I'm going to click on this reupload. And if you have some image like this, you can select this image, and it will automatically crop the image and select the bottom. And now if I reupload and select this image, it will now automatically select the top, as you can see. Now let's click on the generate. And as you can see, guys, we have this result. Now, if you open this image, let's download this image. Let's open this image. And if we zoom in as you can see guys, we have a little bit problem right here. If you have a beard in the model, you will get this kind of result. But let's change the model to recommend it. And this time, let's select this model and click on this generate. Now as you can see, guys, we have this result. Now I'm going to select a different image this time, and now let's click on Generate and let's increase the number of output. And as you can see, guys, now we have these outputs. The results are now pretty amazing. Let's expand this further to a vertical form and here we go. So this is how we can try on different garments onto the model or onto yourself using the AI virtual Tryon feature of Kling AI. 9. KOLORS 1.5 vs KOLORS 1.0: Come back. So we got a new model for image generation called close 1.5. Before this model, we got 1.0. Now I'm going to compare both of these with the same prompt. Now, these examples are for 1.5, and as you can see, guys, with this prompt, we got these results. Now with the exact same prompt, with 1.0, we got these results. Now let's see another example. With this prompt, this is 1.0, which is not bad, but with 1.5, we got these results. Now, the major difference between 1.5 and 1.0 is that 1.5 generate realistic images. On the other hand, 1.0 generates impressive images, but they look like they are generated from AI. And as you can see, this look like an artificial image. But if you look at the result of 1.5, it is more like a realistic to me than 1.0. We have another example. With this prompt, this is 1.5 results. And as you can see guy, the results are looking pretty good. Now with the same prompt, we have 1.0. And as I told you earlier, this more artificial to me than 1.5, because 1.5 is more like a realistic image than 1.0. Now we have another example. With this prompt, we got this result which is not bad. But with the exact same prompt with 1.5, we got these results. Now you tell me which one is good, this one or this one. Obviously, 1.5 looking pretty good. Now with this prompt, we got these results and with the exact same prompt with 1.0, we got this result. With this prompt, with 1.0, we got these results and with 1.5, we got these results. Now, in this example, 1.0 look beautiful, but in 1.5, which I told you earlier, it's pushed the image more to the realism than the imagination. 1.5 push the image to the imagination kind of thing, but 1.5 generates realistic kind of image. Now, if you type something like pure imagination in this prompt, as you can see, guys, this is a pure imagination scenario. With 1.0, we got these results. With 1.5, we got these results. So in this example, 1.0 generates some impressive results because this is pure imagination than the real world. But 1.5 try to generate realistic results, but I think 1.0 win this one. Now with this prom, this is pure imagination scenario. This is 1.0 deserts and with 1.5 with the same exact prompt, this is real that we got. And 1.5 just smash 1.0 because as you can see, guys, it looks like a realistic image. On the other hand, 1.0 look like a concept art. Now we have another example. With this prompt, this is 1.0 results, and with the exact same prompt, this is 1.5. And as you can see guys, 1.5 generate realistic image. Now I try to generate vector art using this prompt, and as you can see guys with 1.0, we got these results, which is not bad. But with 1.5, it generate what I ask him to generate. Because I copy this prompt, I steal this prom from designer domicrosoft.com. AzkaCGuide this is Microsoft designer result. And I copied that prompt and I piece that prompt right here, and we got this result. Now with this prompt, with 1.0, we got these results. And if we go to 1.5, we got these results, which is impressive because if you look at the color combination, and if you look at the graph, 1.5 generate realistic image then 1.0. In 1.0, as you can see, the graph is not aligned correctly, but in 1.5, we got a realistic graph and we got a realistic figure, and we have a good color combination, then 1.0. Now I try to generate some three D animated images. With this prompt. We got 1.0 results. As you can see, these are some creepy results because I ask him to generate three D animation. But with 1.5, we got this beautiful result because 1.5 generate dualistic image. That is why we are getting what we ask him to generate. If you try to generate some anime style pictures, you can also do that. With this prompt, as you can see guide, this is 1.5 results, which is very impressive. In this example, 1.5 look like a finish anime style picture, but 1.0 look like a concept art. We have another example. With this prompt, with 1.0, we got these results, and with the exact same prompt we have 1.5 results. Now with this prompt, we have 1.0 result, and with the exact same prompt, we have 1.5 result. And 1.5 generate what I ask him to generate, I ask him to a man wearing a ski mask sitting on a couch and a big fire happening in the background. And this is what 1.5 generate, and 1.0 just generate these types of images which I really don't like. Another thing that I notice in 1.5 versus 1.0 is that 1.0 contains some artificial softening in these images, but 1.5 generate images without any artificial softening. As you can see, guys, this looks like a cinema camera. And if we go to the 1.0 result, these images have some artificial softening in it. Now we have another example of 1.5. I just change some keywords in my prompt and we got these results of 1.5. Now, as you can see guys in 1.0, again, we have some artificial softening. It looks like that I upscale some low quality image, and this is what the final output is. With this prom, we have 1.5 results, and with the exact same prom, this is 1.0 results. Now let's talk about the vector images. With 1.5 with this prom, we got these results. With 1.0, we got these results, which is impressive than 1.5. Color combination is good. In the background, the red color is good, but on the other hand, 1.5 red color is killing my eyes. Now with this simple prompt, 1.0 generate these images, which is some artificial in it, as I told you earlier. But 1.5 just smash 1.0 in this example as well. Now with this prompt, we have 1.0 results and with the exact same prompt, we have 1.5 results. Now we have another example. With this prompt, I have 1.0 result. With the exact same prompt, we have 1.5 result. Now, 1.5 is looking like that I'm taking a picture of this scene from my camera, in 1.0, but in 1.0, we have this feeling that this is generated from the artificial intelligence. Now we have another example. With this prompt, we have 1.0 result and with the exact same prom, we have 1.5 result. And as you can see, 1.5, try to generate realistic image as much as possible. Now we have another example. This is the prompt, and this is 1.0 result. And with the exact same prom, this is 1.5 result. And as you can see, guys, 1.5 general realistic image, then 1.0. Now we have another example. W 1.0, this is the prompt. These are some results of 1.0, and with the exact same prom, we have 1.5 result. With this simple prom POE of a car, this is 1.5, and with the exact same prom, this is 1.0 result. Now, again, we have some vector image as well. With this prompt, we have 1.0 result, but with the exact same prompt, we have 1.5 result. So 1.5 try to generate realistic vector, but 1.0 have some detail in it. So when we trying to compare both of these model, if you want to generate vector images, then I think 1.0 is the best option. But if you want to simplify your vector, then try to use 1.5 because 1.5 simplify your vector images as much as possible, but 1.0 have some detail in its vector images. Now let's recreate this image using link AI, so I'm going to copy this prompt. These are the result of this prom using Microsoft designer. Now let's paste this prompt and now let's generate image. Now another major difference between 1.0 and 1.5 is that if you want to generate one image using 1.0, it will take your 0.20 credits. But if you switch your model to 1.5, it will take one credit per image. So if you want to generate four images, it will take your four credits. Now let's click to generate with 1.5. Now this is a result of 1.0 with this prompt, and the results are not bad. I'm really impressed with the output of this prom using 1.0. Now with 1.5 with the exact same prompt, we got these results. Which is also not bad, but for some reason, I kind of like 1.0 result. Now, if we go back to the original source, on the other hand, I think 1.5 try to generate realistic image according to this prompt. But 1.0 generates its own version of this prompt. If you want to save your credits, then my recommendation is use one picture, generate both models results, and then if you like one model result, use that model and generate four result if you want to. Because if you generate four images, you will get some different variation. If we compare these four images, we have some tiny differences between these images. This applies same to this 1.5 model. We have tiny changes in four images. And in the final output, it just changed the camera angle. 10. Generate Images With Subject Consistency: Welcome back, everyone. So if you change your model from Cro 1.0 to or 1.5, if you go to text to Image, right here, as you can see, girls, we have a new feature for 1.5 called Upload a reference. So now when you upload a reference, you have to select a subject. Now we have this image as a reference. So to create these kind of images, you have to go to abstodGoogle. Now, once you go to Labs dot Google, you have to select image effects. And now if we open my original image, this is the prom that I use to generate this image using Labs Google image effects. Now, once we upload the reference, we have three options. Either we can select this whole subject, we can select only the phase reference or we can select the entire image. Now you have to keep this in mind, if you select the entire image, then the model will switch from 1.5 to 1.0. So suppose we select entire image. I'm going to confirm, and as you can see, guys, now we have cloud 1.0. And now if we over the mouse onto this small preview of this image, if we click here, this is the edit. If we select the subject phase, and now if I click on this confirmed button, as you can see, the model is now switched from 1.0 to 1.5. If we switch to 1.0, the reference will go on. Now let's upload the reference again. And now this time, I'm going to select the phase. Let's click on Confirm. Now, once you select the phase of your reference, you can type anything that you want this subject to do. Now, if you change your reference strength through 100, then it will copy the exact facial feature of your subject. So let's change this to 50%. And now I'm going to type a boy sitting on a sofa in a studio and let's type dramatic lighting. And as you can see, per image will take two credits with the reference. And if you delete the reference, it will take one credit per image. So we have four variations. That is why we have eight credits. And now let's generate these images. Now with the exact same prompt, let's go to the edit, and now I'm going to select my whole subject. So if you select the phase, it will copy your face as a reference. But if you select the subject, it will copy your entire subject. That includes its clothing, et cetera. Let's click on Confirm, and let's click on generate. As you can see, guys, we got these results. Now as you can see, we got this beautiful result. And if we go back to the original source, as you can see, guys, we have this phase as a reference, and these are a new entire AI generated images. Now we can also change the reference, subject phase reference as well. Now, as you can see, guys, it copied my 45% phase reference of this image. If we change this to 100%, it will copy this exact phase. So let's do that. Let's go back to the phase selected results, and I'm going to change my reference to 100%. Now with the exact same prompt, let's go to edit. And this time, I'm going to select entire image. And now we have 1.0 model. And with the reference, it will take 0.30 credit per image. So let's generate four variation. So the images are now generated. With phase reference, 100% strength, this is what we got with the subject reference as well. Now with only the phase reference, as you can see, guys, we got the result. And as I told you earlier, it will only copy your face. So that is why we have different hairstyle in these four variation. Now we have the entire image reference, and this is what we got with 1.0. Now let's select another subject. I'm going to delete this subject. Now if we go back to labs dot Google Image effects with this prompt, I generated this image. So let's download this image. Now I'm going to select this subject, click on Confirm and with the exact same prompt. With 100% phase reference, we can generate let's let this aspect to show. And now let's generate these for variation. Now as you can see, guys, we got these images. Let's change the subject reference to 100% as well, and I'm going to change my pace reference to 60% because I want AI to take the reference of the subject, not too much, and I want the clothing of the subject 100%. So that is why I change my strand to 100%. Let's generate these results. Now as you can see, guys, we got these results. Now we are getting a different phase because I type a boy and let's change the keyword to men and now let's generate. So as you can see, guys, in the new generated results, it takes 100% reference of my subject. As you can see guys, if we go to the original source reference, let's go to edit. We have this jacket, and if we go to the generated images, we got that same jacket. And now as you can see guys, we got these results. So the reason why this reference image feature is important is because you can now generate consistent images, and with that consistent images, you can also turn your images into videos. So if we go back to the animated corrector, as you can see, guys, we can take this boy to anywhere. So we can also type sitting on a sofa to sitting in a car. Now with this prompt, it will generate images, and now we can turn that images into videos. This way, you can create consistent videos with that same subject consistency. So now, as you can see, we have that same subject in a car. We can change the sitting in a car to driving a car. Let's add a comma with dramatic lighting. Let's turn right. And now, as you can see, we got these results. So now we can create consistent character videos using this reference feature of ling AI. 11. KOLORS 2.0: Welcome back, everyone. So Cling AI launch their new image model, which is called Colors 2.0. Now to access to Color 2.0, you have to go to the homepage of Cling AI, and by the way, this is a new interface. We can select image, and we can switch the model from here. I'm going to select Colors 2.0, which is the latest model. Now we can type in any prompt. You can go to Chat GBT and you can ask him for some ideas. Like in this prompt, I gave me ideas on Mini World for video generation, and I got these prompts for Mini World. Now we can copy any prompt. So I'm going to copy this prompt. Now you can test this prompt before you paste that prompt in cling AI, go to image effects of Labsto Google, and you can paste this prompt. This is a free image generation with unlimited credits, so you won't lose any credit. So you can test your prompt here first, and then you can try this prompt later in cling AI model. So you won't lose any credits. Now, as you can see, we got this beautiful world. Now we can paste this prompt. Now you can change your aspect ratio from here and you can also change the outputs of your prompt. Now you can change the model from high resolution to standard. I'm going to select high resolution. Let's generate four, and let's generate a standard, so we can compare them later. So the images are now generated. Let's look at the high resolution first. So if you open an image, as you can see, we have this new interface of viewing an image. These are the height resolution result of clers 2.0. Now we have these standard images, which is also not bad, but the height resolution have some depth of field in it. You can also upscale your image as well. You can expand your image as well. This is another prompt that I try with colors 2.0 high resolution. We have this result. I have this regular soda gain superpower like flying laser phis, and with 1.5, we have this result. We have colors 2.0 result right here, and we have this example, which is looking realistic. And we have this example as well. Now if you want to only view the images, then you can switch from all to images. And now, as you can see, now we have only the images. And we have some other prompt as well. Like in this prompt, we have these examples. As you can see, the images with colors 2.0 are looking crazy. We have this example. Check out the details. And with this prompt, we have these images. And with this prompt, we got these crazy images. M let's compare the colors all the model. As you can see, we have colors 1.0 with same prompt. If I open this image, this is the result that we got using this prompt. Now with color 1.5, we got this result. And with the same prompt, with colors 2.0 high resolution, we got these results. This is not what I was expecting from colors 2.0, but this is what it is. And now with colors 1.0, with this prompt, we got these results and with the exact same prompt with 1.5, this is what we go. And with colors 2.0 high resolution, we have this insane upgrade. I mean, it looks like a real life picture. Now we have this prompt, and with colors 1.0, this is what we got. And with the exact same prompt, this is 1.5 images. And with 2.0, we got these images. I mean, in this case, I think 1.5 was look realistic to me than 2.0. 2.0 overall look fantastic. I think we have some art here. We have five finger and we have a thumb. So in this prompt, I think 1.5 is the clear winner. In some cases, 1.5 perform better. But if you look at overall, 2.0 is beating the 1.5 and 1.0 in average scenario. Now we have this prompt with 1.0, we have these images, and with the same prompt, we have 1.5 images. Now with the exact same prompt, we have this image, and 2.0 perform incredibly better than all the other models. As you can see, we have shallow Daptofeld. We have a cinematic scene. You can even generate this image into a video, and it will look fantastic. Now we have this prompt with 1.0. This is what we got. And with 1.5, we have a little bit improvement in these images. But if I show you 2.0 result, you will be surprised. I mean, check out these images. They look like a real life pictures to me. Now we have this prompt with 1.0, we have these images. And with the exact same prompt, this is 1.5. 1.5 give me the images that I want from this prompt. And with the same prompt, we have 2.0 result. I mean, everything look realistic except the birds. I think 1.5 perform better in this prompt. It gives me the bird shape that I want, but everything else, if you ignore the birds, I think everything else is look better in 2.0. We have this prompt with 1.0, we got these images, and with same prompt, we have 1.5 result, which is a huge jump from here to here. And with the same prompt, we have 2.0 result. In this prompt, I also like 1.5 result better than 2.0. So as I told you earlier, 1.5 perform better in some cases than 2.0. Now we have this prompt, and this is 1.0. And with the same prompt, we have 1.5. As you can see, we have a little bit improvement in the image. And with the same prompt, this is 2.0, and there is a huge jump from 1.5 to 2.0. It look realistic. But again, I think, in this case, 1.5 also perform a little bit better than 2.0. Because it also look like an artificial image. But if you look at 1.5 image, the image look nice and I don't seem to notice any error in the image. So in this prom, 1.5 is also again a winner. Now we have this prom, and again, we have tomatoes again in the image. With 1.0, we have these images, which is also not bad. And with 1.5, we have these images. And with colors 2.0, we have this insane image. And clearly, as you can see, we have a winner right here. 2.0 perform better in these images than 1.5 and 1.0. Now we have this example, and before I show you 1.0 and 1.5 result, this is the image, and this is 2.0 result. As you can see, it looks like a game picture, and with the same prom, this is 1.0, and this is 1.5. And I don't think this prompt work in all the models because all the model give me the concept art except 2.0, give me a game like picture. Now we have this prompt and we have 1.0 images, and I really impressed with 1.0 as well. It performed better in this image. And with 1.5, we have this insane upgrade. As you can see, from here to here is a huge upgrade. And with the same prompt, we have 2.0, and as you can see, 2.0 perform better than all the other models. Now we also have a upgrade to view the image. As you can see, right here, we have a small thumbnail of all the generated images. We can view them by scrolling down or scrolling up like this. Now, if you open an image, we can also use the scroll to view the image as well. You can scroll down or you can scroll up like this. Now you can also use the keyboard arrow down key to view the image to downward, or you can use the upward arrow key to view the images to upward. 12. Restyle Images Like Ghibli Style: Welcome back, everyone. So Kling AI image model Colors 2.0 offers a new feature called restyle. You can restyle your image into any style that you want. I input this image, and I type in the style of 90s anime. And this is what we got. So to restyle your image, first of all, all you need, you can switch your model to colors 2.0. And you can upload your reference. Now, once you upload your reference, you can either select the style. I'm going to select this style, and by default, we have the standard mode, and you can also change your outputs. So I'm going to select two outputs. Now you can also click here and you can refresh the hints that the cling AI offers. And as you can see, this time we got the style. So I'm going to select Pizle art and let's generate four images again. Again, we can refresh, and this time, I'm going to select comic and let's generate four images again. And as you can see, we have these images. They not look fantastic, but this is what we got. Now with the pixel art style, this is what we got. And I really like the pixel art style. And with the comic style, we got these images and I really impressed by the comic style and with the pixel style. With this style, we have so many errors in the image. Let's try some other style. Let's select the anime style, and I'm going to also select three D style. Now with the three D cartoon style, this is what we got. And with the change to anime cartoon style, this is what we got. So this is how you can use the restyle feature of the new model of colors 2.0 of King AI. 13. Image Editing: Welcome back, everyone. So cling AI New update gives you a new feature, which is called image editing. If you go to the ling AI homepage, and as you can see, right here, we have image editing. If you click on it, we have this interface. Now this is how it works. You have to upload an image. Now my image is uploading. Now, once your image is upload, this is the interface of image editing in painting. Now as you can see, we can use the scroll to move this image. Now I can use my scroll button to move this image. We can press Control, hold it, use the scroll to zoom in and zoom out the image. You can also change your screen from here. Now, these are all the shortcut keys that are available for this interface. Now, right here, as you can see, we have some tool. We have brush selection. So basically, first of all, we can increase or we can lower the brush size from here, and now I can select brush selection. Now I'm going to select my watch and I'm going to type a prompt. Roll golden watch, and we can select the output. So I'm going to select two outputs. And as you can see, if you click here, you can increase intensity if you want the effect to stronger or if you want the effect to lower. So I'm going to set my effect to the medium. You can also inverse your selection by clicking here and you can also clear your selection, and you can also undo or redo this action. Now we can click here to paint this effect. And once you submit your selection, we have this window that appears that it is generating the effect that we ask him to generate. Now you can click here to continue editing. You can again ask any changes in your image as well. Now we have three selection. We have Bush selection, we have box selection, and we have quick selection. Now I'm going to clear my selection, and I'm going to also clear my prompt. First of all, let's look at the result. And as you can see, we have a Golden Rolex watch right here. Now you can also use the same selection by clicking onto this image. And if you want a refresh start, then you can clear your selection. We have the bred selection, and we have other two selection. So you can simply select the box selection, and again, if you want to change the watch, you can simply select the watch. And if you want to select this part, you don't have to undo this selection. You can also draw another area like this. And now we can this time change this to caso modern watch. As you can see, it is not generating the effect. Now we have a quick selection. I'm going to clear my selection and I'm going to select quick selection. First, it will analyze your image. Once it analyze your image, as you can see, we can select any part of the image to change that part. We can also select the background or we can select the shirt as well. I'm going to select my shirt, and it also selected the beard. So we can also erase this part by clicking on this eraser tool, increase the brush size, we can erase this selection. And now we have this. Now I can change my shirt, white shirt. So while it is generating the shirt, let's look at the watch, and as you can see, we have this Cassie watch on my wrist. As you can see, it does not change my shirt. So again, we have a little bit changes in the shirt, but it does not change the shirt to a pure white. So we can select this input again and we can ask him again to change the color of the shirt to plain white. So this time we're asking him properly to change the color of the shirt. Now, while the shirt color is changing, we can clear the selection and we can brush the subject like this, increase the brush size, and we can select the whole subject like this. Or the reason why I'm doing this is I'm going to show you how we can use inverse selection the right way. If you want to remove or if you want to change your background, simply select your subject first and then inverse your selection. So this way, you don't have to paint your whole background. So we can just paint a little bit, not too much. The edges of the subject. Let's erase this part. And now I can ask him to change the background to a studio, match light and shadow of this atmosphere field. So let's ask him to change the background to a studio, match light and shadow of this atmosphere. So while it is generating the background, let's look at the result. And again, we don't have the white shirt. So let's ask him change the shirt color. So this time, I ask him to change the color of the shirt to plain black. While it is again changing the shirt color, let's look at the background. So as you can see, it changes my background. And if you look at the shirt result, the shirt stays the same. So I think we have to manually brush the shirt so I can clear my selection, increase the brush size, and we can paint the shirt. Now you can also change the color of the hair as well. So I'm going to select my hair and let's change the color of hair to red. Now, this time, as you can see, we have the plain black shirt. So the reason why we were getting the error is because we did not select the shirt properly. So again, this time, I'm going to ask him to change the color of the shirt to plain white, and hopefully we will get the result that we want. Now, as you can see, it changes the color of my hair, but it also changed the hair cut as well. So I'm going to select this input. I'm going to clear my selection, and I'm going to draw this hand, and I'm going to also draw this area of my hand. So I have this prom arrow tattoo on both hands. Now, this time we have this blue shirt, and finally, we have the white shirt. So in order to modify an object or subject, you have to properly select that area. Now, as you can see, we have this arrow tattoo on my hand checkout. We have another example. If I open this image, as you can see, I also changes the background of myself. In this example, I have the studio background. If you go to the in paint, replace background that match the lighting of the scene with the subject, add a studio. And as you can see, we got this result. And in this example, we have replaced the background that match the lighting of the scene with subject. Add a studio background with medieval stuff, and this is what we got. And in this example, I selected the beard and my hair, changed hair and weird color to blue, and this is what we got. And this time, I changed the hair color to blue only, and we got these images. And in this example, I change the shirt color to plain black, and I use box selection. Now, if you notice I also select my face as well. That is why we got this weird face effect right here. So if you avoid this selection, it will change the color of the shirt properly. So in this example, I also use a quick selection to select my shirt. That is why it did not change the color of my shirt to red. As you can see, we are not getting the color of red shirt. In this example, we have these changes. Now you can also select multiple object and you can change them. So in this example, I selected my belt, my watch, and my glasses, and I typed this prompt. Change watch to casseo, change glasses to sunglasses, change the belt to brown. And as you can see, we have the wa change belt, and we have sunglasses as well. So you can select multiple objects in one image at once. You don't have to select individual object. You can select multiple object at once. And in this example, I change my background to empty room, and this is what we got. I again paint my whole background, including myself, and I typed this prom, clear the background, make it an empty off white room. And as you can see, we have these images. If you don't want the error in your image, then you can use Quick Selection to select your shirt or any object, and then you can use the bras selection to select the eddies because as you can see the quick selection, we select the object, but it won't select the edges of that object. In this example, it select my shirt, but it is not selecting the edges of my shirt. That is why we are getting this kind of error. And that is why we were getting the error. So manually brush the edges, and hopefully you will get the result that you want. Image editing, we have another feature called expand. You can expand your image as well. So in this example, I select a free expand area, so I selected this area, and as you can see, without any prompt, it gives me these images. And if you type a prom, so for example, I selected this area as well, and I type abandoned hallway with scratches and graffiti on wall and check out the result. Now we can go to the Expand. We can select any image. So I'm going to select this image and go to Expand. We can use the original or we can use a custom ratio. So I'm going to select this ratio, and I'm going to also place my image right here. And without any prompt, let's expand the background, and let's type a prompt as well. So without any prompt, we have this image. Now, I really like these two images except this one and this one because in this one, I think we can really see the merge effect right here. Now I type this prompt studio lights and plants plays in background as you can see, we have the plant and we have studio light right here. So it beautifully merges these things and place them to expand the background. I really like this one and this one because as you can see in this plant, we have this light effect as well. So this is how you can expand your background, and this is how you can use the image editing of Kling Ai 14. Basic Introduction to Generate AI Video: Welcome back. Now you have to click here and go to the videos in order to generate clips using cling AI. Now, there is a few ways to generate clips using Cling AI. The first one is text to Video. In text to video, you have to type a prompt in order to generate clip. So now let's do that. Now I have this prompt. You can copy this exact same prompt in the settings. As you can see, guys, we have to slider. We have creativity and we have relevance. Now in the mode, we have standard and we have professional. If you are using the free account, you will only get the standard mode. So now let's check the standard mode for this example. Now we have the length five to 10 seconds. We have the aspect ratio that you can change as well. You can select any aspect ratio that you want. Now we have number of generation. Right now, I have set to one video generation. Now below the aspect ratio, we have camera moments. If you click here, we have so many camera moments. Now below the camera moments we have negative prom. So what is the negative prom? Negative prom is something that you don't want in your generate video. For example, if you don't want blur in your video, just type the blur. If you don't want distortion or something like frame infringement, you can type that word here and it won't generate that thing in your video. Now let's click on this generate button. So I also use Microsoft Designer for this prompt. As you can see, guys, this is the result of this prompt in Microsoft Designer. Now I am thinking that we will have a different result than this one, because this was generated in Microsoft Designer, and we are using the King Eye, which is totally two different kind of tools. While the video is generating on the right side, we will have the results of generated videos. As you can see, guys, I generate so many videos. You can click here to change your videos to text to video. So these are all my text to video. You click here, you can change this to image to video. These are all my images turned into AI videos, and you can also select all video, you will be having all the videos displayed on the right side. You can also click here. You can also change the view from simple to detail view. Now, in detail view, as you can see, this is the new video that I am generating using this tool. These are all my old videos. In detail view, you will have the prom will have the reference image. If you have one, in this example, I have the reference image, so these are all my reference images. It will also display the version that you used to generate that video. So for example, for this one, I use Kling 1.0. If I scroll down for some videos, I use Kling 1.5, like this example. Let's change that. You can also click A to select any video or picture you can download. Or you can delete that video or picture. You can also set that to your favorite, or you can click again to unfavorite, that thing. Now let's click on it again. Now the video is generated. Let's view the result. And it is totally different from the Microsoft designer, as I said earlier. So this is how you can generate a clip by using a prompt in text to video using ing AI. Now, in the next lecture, we will talk about what's the difference between the standard mode and the professional mode. 15. Generate Impressive Text to Video: Come back. Now let's copy this prompt. I used Microsoft designer to enhance my prompt, and I got these results. Now let's go to the King AI and paste this prompt. Now for this video, let's select the mode to standard mode, and I don't want to tweak any setting. And now let's click on the Generate button. While this video is generating, we can select the professional mode. By the way, if you are using the premium account, then it will cost you 35 credit to use the professional mode. Now we can generate this prom, so the videos are now generated. So this is the standard mode video. It's looking pretty good. And this is a professional mode. As you can see guy, there is a huge difference. There is no frame infringement. There is nothing wrong in this video. It looks like a real footage from the real life. Now let's change the cling 1.0 to 1.5. Now when you select 1.5, the camera moment will disappear. And we can't change the professional mode to standard. So now let's generate. Now there is a huge difference between professional mode and the standard mode. The professional mode will stay consistent and generate your video. There is nothing wrong. There is no frame infringement in this clip. But the standard mode just generate your clip, whatever you ask and it will stay inconsistent. As you can see, guys, we have something frame infringement in the foreground, and we have something weird dots or ringement appearing in the mountain, as well. So if you want to generate high definition results, then use a professional mode, but you have to upgrade to the pro version. Now, by the way, if you are revisiting your generated videos and you want to view which mode I use and which setting I use to generate that clip, then you can also view that. For example, if I select this video. As you can see, guys, I changed the mode to standard. But if I select this video, it automatically changed the mode to profession because this was the mode I select when I was generating this video. Not only this, if you select any camera control, then it will appear right here. If you change any aspect ratio, length, et cetera, then it will appear to the left side. Now the video is regenerated, here we have the result. I kind of like this one because it generates what I asked for. For the professional mode, I think I like this one, but the new version is looking more beautiful and better than the previous ones. Now, if I also regenerate this video, we will get different result. So why not do that? Now, if you are not getting the result that you want, then regenerate your result. So in this example, we get a different result and better than the previous. This video that we generated using one but five is not bad. But the new video is looking more beautiful than the previous. So now we have the regenerated video result using 1.0 perfection mode. If you want to generate videos, then you have to try version 1.0 professional mode version and 1.5 perf mode version. Because you will get different results, then you can choose video you want to use in your projects. I can use all these videos except the standard board because the standard mode is not useful. You can use in your YouTube videos, but it is not useful because we have inconsistent video, whereas the pfession mode 1.5 mode is much better way to generate videos. 16. Creativity vs Relevance: Welcome back. Now I will show you what's the difference between creativity and relevance. So for this example, I use this prompt and I generate these four results. Then I use this image as a reference. Then I generate these four results. Now let's click here to bring this image to life. Let's close this tab. Now if we just stay here in the center, it will just generate a normal video, milking, pouring on this chocolate. Let's change this to creativity and let's generate. Now again, I'm going to change my slider to the right side to the relevance. And now let's click on this Gnute button. Now the results are generated. Here we have the creativity result. And here we have the relevance result. So in creativity, as you can see, guys, it's just adding something on its own. Whereas in the relevance, it just stick to the topic and generate what we ask for him to generate. Now, what if we just don't change the slider to the extreme relevance or extreme creativity? Let's change the slider to about 0.35 creativity. And this time, I am going to select the standard mode, and now let's generate this clip. Now let's change the slider to about seven relevance and generate the clip. Now let's change the ling 1.0 to 1.5. Now let's change the slider to creativity. And again, change the slider to relevance and now let's generate. So now we are generating the four video at the same time. So the results is now generated. So if we just change the slide a little bit to the left side to the creativity, we have this result. And if we change the slide to the right side a little bit to the relevance, we have this result. Now I know these two result is looking similar, but there is a huge difference. So in the relevance, stick to the topic. It's gather the milk onto this side, where in creativity, as you can see, guys, the milk is spreading to this portion, not just gather to this side. In the relevance, as you can see, guys, the milk is now gathered to this side. Now let's take out the result to the extreme creativity and relevance of 1.5. This is the extreme creativity, 1.5. The results are not bad. Now let's check out the dileman result. Now if we change the slide to creativity is 0.3, I think we will get something more better result than this. So now let's generate clip. While the video is generating, let's take out some other example. So in this example, I use this prompt. I use 1.5, and I change my slider to extreme creativity. So we have this result. And in next example, I change my slider to extreme relevance using 1.5, and we got this result. And if we just don't change the slider at all, we have this result. And in this example, I just use an image. Without changing the slider, we got this result. So now the results are generated. If you use just a little bit creativity, you will get some interesting results. And I'm using the standard mode for this one, not using the professional mode. And I'm using latest 1.5. Creativity is set to 0.3, and this is the worst result that we have ever generated using this image. So my final words, don't use this slider above 0.65 and below up to 0.35. If you want to generate something interesting, then don't push this slider to extreme left or extreme right. And this is the difference between creativity and relevance. 17. AI Camera Movement: Welcome back. Now go to Microsoft Designer, and I want to create a chocolate commercial. So I'm going to write Chocolate Commercial in a hotel. Now we can enhance the prompt. I think this prompt is good. Let's copy this prompt. Go to Kling AI, and we can paste this prompt. Now if we scroll down, you can see guide the camera moment is disappeared. So to use the camera moment, we have to change the Kling 1.5 to Kling 1.0. Now we have this camera moment control. If we click here, you can see we have different kinds of camera moments. If we scroll down, we have also master shot. You can select this and you will get different camera moments. Now I will explain every camera movement. So first of all, we have horizontal. Now we can change the slider and we will get different horizontal position. If I change this to the left side, I will get a horizontal moment from center to the left side, and if I change this to the right side, then we will have a camera moment from center to the right side. Instead of explaining, I'm going to show you. So first of all, let's select the horizontal and let's change it to the left side. So in this prompt, we have a chocolate commercial, and in the commercial, the camera will change the position to the horizontal position from center to the left side horizontal. So now let's select 1.0, change the motor professional, and let's generate clip. Now the video is generated, this is the result. So as you can see, guys, in this video, the camera is changing the position to the left horizontal side. Which we set. Now we can change the camera moment very easily. You have to just change the slider to where you want to set your camera moment. So let's try out the Zoom out camera moment. Now I have generate different camera moments. So this is the Zoom in extreme Zoom in camera moment. And this is the pan 3.4. This is a tilt five and this is extreme role to minus ten. As you notice, guys, the videos that we generated using these camera moments are inconsistent. So all the clips are inconsistent. There is something happening in the video in the background, in the foreground while you use the camera moments. There is a subtle camera moment that you can use and you will get less this infringement. But if you use extreme camera moments like we use in Zoomin and in extreme minus ten horizontal, you will get extreme infringement. Now let's reset the camera moment. Let's generate the simple video. Now I am going to copy this prompt. These are the results of this prompt. And now let's go to the video. Let's paste this prompt. First of all, let's generate without any camera moment. So without any camera moment, this is the result. And as you can see guy, the video is consistent. There is no weirdness happening in the video, and this is the three D character video without any camera moment. But as you can see guy, the results are pretty amazing. So for this video, we have the same prompt which we use in three D character, and I use master shot, move left and zoom in. And here we have the result. As you can see guy, the result is inconsistent. There is some weirdness happening in the shot. We have another example, master shot, move forward and zoom up. There is not much detail in the video, so that is why we have this smooth result. And we have another example of master shot, move right and zoom in. Again, inconsistency in the video. So I don't think right now we can generate consistent result using the camera moments. If you want to experiment with the camera moment, you can do that as well. With the free version, you can just select your standard mode and you can use any camera moment. And I don't think that you will lose any additional credit for using the camera moment. But if you want to generate consistent result, then I recommend don't use any camera moment because right now the camera moment is not usable. 18. Negative Prompt: Welcome back. Now, in this tool, we have another amazing option called negative prompt. So if you don't want to see anything in your video, just type that word, and it will generate without that thing that you type here. So, for example, in this video, as you can see, guys, we have this weird distortion. So in this example, I type infringement and motion blur that I don't want to see in my video, and here we have the result. As you can see, guy, there is no distortion in our video. We have another example where I type just infringement. So we have this smooth transition between two frame. Whereas if I play this video, we have this weird infringement. Like this one. In this example as well, I just type infringement and we have this smooth transition result. By the way, if you are using the standard mode, you can also get some amazing result. Like this one, I use my standard mode and I select this master shot. And I just type disfigurement, motion blur and infringement. And here we have the result. There is a little bit infringement, but not a lot. We have this result as well, and we have this one as well. We have another one For this one, I use the standard mode, and we got this result. As you can see, guys, the character is colosing its eye, and we have something distortion happening in the short. So to avoid this distortion, I just type distortion in the negative prompt. And as you can see, guys, now we have this smooth result. And by the way, I use the standard mode, not the professional mode. As you can see, guys, the ten credit is appearing in the generate So by typing negative prompt, anything that you don't want to see in your video, you will get some stunning result, even if you use the standard mode. And after I regenerate this again with this camera moment, with these negative keywords that I don't want to see in my video, I got this result, and there is almost none infringement in the video. So this is how you can generate stunning result, typing negative prompt using ling AI. 19. Generate Image to AI Video: Welcome back. So the next way to generate videos is to use images. Go to the image to video. We can upload the image or we can go to the AI Images, select any image. You can choose any image, and then you can click here, bring to life. You can generate video using your image. So we have already done this, so let's delete this. Now go to the Microsoft designer. You can type any prompt that you want, enhance your prompt. You can select your size, what the size of your image would be. You can select any size. And now let's click on this Gen wait button. Now, here we have the result. You can select any image that you like. So I kind of like this one. You can click here to download any image that you want. Now go back to Cling Eye, go to the image to video, click here, select that image that we just download. Click on this open. You can drag and drop your image as well. So now we can drag and drop this image as well. You can type any prompt is optional. There is some other settings as well. And for this example, let's select the standard mode. Now we can click to generate this video. So the video is now generated. Here we have the result. You can bring any image to life. You can upload any image. It doesn't matter. You can upload yourself as well. Now in this example, I upload my image, and here we have the result. Not only this, you can also bring your old photo to life. You can also upload your manipulated photo. As you can see, guys, I photoshop myself in this picture, and I bring this picture to life as well. This is another example as well. We have this example as well. And we have this example as well. So you can see, guys, it's pretty amazing how you can bring your childhood photo to life. Not only this, you can also tell him to do anything that you want with your photo. So in this example, I type knife cutting orange into two pieces. And now here we have the result. Knife is now cutting the orange. And in this example, I ask him make this video set in a cyberpunk style. And this is what I got. Pretty amazing, right? So this is how you can generate your videos using the image. 20. Start and End Frame: Welcome back. So in Killing A, we have another amazing feature called add ndframe. If we enable it, you can see we have two windows to upload the images. This is the start frame, and this is the end frame. Now go to the Microsoft designer. You can type this prompt. You can generate these images. For this example, I'm going to download this image. Now you have to select your end frame according to the perspective of your first frame in order to achieve stunning results. So I think Number four image is close to the orange image. So let's download this as well. Now we can click here to upload the start frame. Now let's click here to upload the last frame. And for this example, I'm going to select this ten mode, and now let's click on this ten Wt button. So here we have the result. As you can see, guys, the orange is now transforming into Apple. Now I also auto extend this video, and now we have this generated video. Then I also extend that video five second further with the prompt hammer mashing apple, and now we got this result. You can also ask him to do anything. In this example, I type Apple transforming camera stay static. So the camera won't move, the camera will stay in the position. There is two different perspective. That is why we got this weird result. But if you got the perspective right and the table position is the same, then you will get a stunning transformation video. In this example, I just type Apple transforming. Before we got the orange at the start frame, now we have the Apple at the start frame and the orange at the end frame. You can do different types of crazy thing using the end frame feature of this tool. Let me show you some few examples. So we have this example. This is my start frame, and this is my end frame. I typed the pro car crash through the wall and stuck between two walls. Chain my slider to 0.8 relevance and use standard mode for this generation. In this example, I use this frame and add myself into the venom, and then I use this at the end frame. As you can see, guys, and I type man becoming venom. I didn't change any settings, use the standard mode for this generation as well, and I got this result. After I generated that video, then I improve some lighting on my face, add the venom face at the start frame and add my face at the end frame, and type a symbiote is revealing a man inside it. Use the Dan mode for this one, as well, and I got this crazy result. After that, I regenerate again my video, and I got this crazy result. So you will get different kind of result if you regenerate. I also use my two pictures at the same location, but with different angles I also use the standard mode for this generation as well, and this is the professional mode result. As you can see, guys, my face is now stay consistent. And now we have this example. If you have the same image with the same perspective, you will get this kind of result. As you can see, I type camera stay static and we get this crazy result. Now if you use a professional mode with the add end frame feature, you will get consistent result. So this is the standard mode video. And as you can see, guys, my face is not visible properly, and this is the result with a professional mode. The video will stay consistent. We have these example as well. And this is another example of end frame feature of Kling AI with a professional mode. And we have another example with a professional mode with these two images. As you can see guy, the results are looking realistic, and we have this example as well. With the frame switch, you will get this effect. Now, you don't have to use the first frame and the last frame in this combination where we have a white angle and we have a close up angle because you will get something like this. You will get extreme fringing effect like this. And we also have this example. And as you can see guide, we got some pretty amazing lasers. Even the people that are in the background are moving very correctly. So you can create anything with this switch. You have to just add a star frame and you have to add an end frame, and it will create something in between those frame. 21. Turn Yourself into Iron Man: Welcome back. Now, even if you use the free version of Killing AI with the standard mode, you can create something like this. So without any special skill, you can create stunning visual effects as well. And if you use a professional mode, then you could get some stunning results as well. Now I'm going to show you how we can create something like this. First of all, move to the photob.com. This is a free alternative to Photoshop. We have two screenshot of Ironman. First of all, uncheck Autoselect and then check Transform Control. By enabling the transform Control, you will get transform Control. We can press Control C to copy this layer. Go to the Iron Man press Control V to paste yourself. Now we can convert this layer to a smart object. Now press Control and hold them. Use your mouse scroll to zoom out. And now we can resize this person or the subject, press Inter, and we can again press Control plus or hold them, use scroll to zoom in. Now I'm going to lower my opacity and now we can align the phase with the ironman phase. Now we can change the opacity to about 100%. Click here. And now let's align this right here. Now I'm going to enable raster mask, click here, select your layer first, and then click here. Now we can click and hold it and we can select the paint bugget. We can swap the color as well, change the color to black, and now let's click select the mask and click. Now we can select the brush, press right click. We can change the size of the brush right here and now we can swap the color. So click here. Now we have the white color. So now we can again use Control and t, hold them, use the scroll to zoom in. Select the raster mask and now we can paint. Press Kontozi to undo. Right click, and we can lower the size of the brush. And now I can brush right here. So you have to carefully brush. Inside the I did man mask. Now we are done. We can select the move tool press control or Zoom out. Now we're almost done. Now I'm going to click here and I'm going to add this adjustment layer, right click on the adjustment layer and create clipping mask. So it will only affect the phase, not the whole thing. First, I am going to increase my contrast, and I'm going to lower my brightness. We can turn this on and off. Let's hide this layer to get the color right. Let's lower the brightness a little bit. I'm going to also add a selective color. Now if you don't make this eclipsing mask, then if we select the neutral color, it will affect the whole image. As you can see, guys, it is now effecting the whole image. And if I change this to e clipping mask, it will only affect the layer which is underneath this layer. Now we can reset this layer and we can match the color with the iron man. Now I think we are done. Let's place this to about. Now we are done. Now the last thing that we need to do is we can right click off the person face, go to the Blending option, unable inner shadow. Go to blend mode, change the blend mode to normal, click here and change the color to black. Now we can increase the size, increase the opacity. Let's lower the past a bit. Now we are done. Now go to this screenshot, click here, press Control C to copy this and press Control V. Now we can place this screenshot underneath all the layers. Now we can select all this layer, select this layer, press Shift and select the first layer. So it will select all the layers. Right click and convert this to a smart object. Select this and right click and also convert this to a smart object. Now we can click here and we can also press Control and select this one. Right click, convert this to a smart object. Now we can go to the filter, go to the noise, add some noise. So I'm going to add 5% noise. Maybe change this to a 3%. Now click here on the filter again, go to the sharpness and select smart sharpness. Click Okay. And now we are done. You can also color grade your pictures if you want to. So I'm going to just add something green and add something like Merchant. And if we turn this on and off, you can see, guys, we have some color grading which will help us to match the face with the scene. Now we can click here on the file, go to the Export and save this file as JPEC. We can change the quality to 100% and now click on the save button. Now select this layer, double click. We can turn this layer off. Now we have this layer, we can press control as to save this and now we have this update version. Now we can go to the file and we can save this as JPA. So the reason behind why I add this screenshot underneath this screenshot is to make them one layer, smart object. So then we can color grade and we can add some effect on top of this layer and it will effect all the layers which are underneath the main layers. And now go to the link AI, we can delete these images, and we can upload the new images. Now you can use the standard mode as well if you want to and we can click on the Create button. The videos is now generated. Now, if I take this as a first frame and take this as a last frame, with the standard mode, this is the result. And if we set this first frame and set this as a last frame, then we got this result with the standard mode as well. So I read and read this video with the professional mode and we got this amazing result. This look more realistic than the other generated videos. By using this technique, I also created myself as a venom. Now, with the help of AI, you can turn yourself into any superhero without any special skill. 22. Motion Brush: Welcome back. Now we talk about the motion brush, and this feature is a game changer for the AI video generation industry. First, we have to upload our image. Now, once you upload your image, you have to change cling 1.5 to 1.0 because this feature is not available in 1.5 right now. Now we have to click here. And as you can see, guys, we have this simple guide. So let me show you how it's worked. First, we have this brush. We can draw something, we can draw anything that you want to move. And we have eraser. If you want to erase your brush, for example, if I draw this rock, and if I want to erase my brush, I have to select the eraser, and then I have to remove the rock. I can also increase the size of the brush as well. Now we have auto segmentation. So right now, as you can see, guys, I'm just brushing myself. But if I reset, you can also reset your brush area. I can automatically select myself by enabling the auto segmentation. Now I'm going to select my hair, my glasses, my pants, and my shoes as well. Now I can uncheck this and I can brush the missing pieces. So we will get the accurate result. Now we are done. So at the right side, we have the track, and this is for this area, the green area. So right now, I'm going to select my track, and we have to draw the direction where I want to move myself. So I want to move myself to this direction, right? Let's delete this brush, and let's draw to this side. Now you can also delete your path, your track as well. You have to click here on the cross to delete your path. Now we will get different areas. You can select multiple areas to move, but right now I have only myself in this picture, so this is enough. Now, at the end, we have this add static area. So the static area won't move. So for this, I have to select everything that I don't want to move in my picture. I think this is enough. Now, right here, as you can see guys, if you select something that you want to move, but you accidentally select that as a static area, then you can also undo that. As you can see now this area is now clear. You can also redo as well. Or you can reset everything if you want. Now we are done. We can confirm and we are done. And by the way, I'm going to reopen the motion brush again. For example, if you don't draw any drag, if you don't draw any path and you click on the confirmed button, we have this added message, not drawn, not paired. Motion brushes won't effect submit. So you submit if you confirm this, if you submit this brush area, it won't affect anything because you did not draw the direction. So in order to move something in your image, you have to draw that thing and you have to draw a path. If you did not draw any path, then it won't move that object or a person or a thing. So now let's confirm. And now let's use the standard mode for this and we can generate this video. And now here we have the result. As you can see, guys, if we use the professional mode, we will get some consistent result. Right now, it's not consistent, but it's looking good. Now we don't have to just go back and click on the video in order to generate videos from the image. We can also delete this. Don't worry, it won't delete your video. And now for this example, I'm going to click on the motion brush. So let him analyze your image first, and now we can enable segmentation. I can select myself. Let's uncheck. Now let's select the track and let's draw a path right here. So now I'm going to select the static area, enable auto segmentation. And we are done. Let's select this person as well. Now, click on the confirm and generate. So now we have this video. So sometime it generates something that you did not expect him to generate. Don't worry, I think we can fix this. So while this video is generating, we can delete the image. We can upload another image. And in this example, we will draw multiple objects in order to move them. Now the image is uploaded, click on the draw motion. Let him analyze the image. We can enable auto segmentation, select the idea number one. We can draw this got, select the area number two, select this one. Now I'm going to select Track two, and let's move this go to this position, select the Track one, and move this go to this position. Now we have to draw background as a static area, so the background won't move. Now we are done. We can click on the Confirm, and let's generate this clip. And I have this example for you guys, and I want to explain something. Let him analyze the image first. So for example, let's take automatic segmentation. If you click the background, if you draw the background as the motion area, the moment area, the moment path, if you select the static area, you won't be able to draw the background as the static area because we draw the background as the movement path. So we can reset. Let's reset this again. And for example, if you draw your character as a static path, then if for example, the area number one, I have to draw my face, then I won't be able to draw my face because this area is now set to the static area. So make sure to brush very carefully. And in this case, for example, if I select this portion, this part of the character, it will select all the background as well. So enable to draw this area, you have to manually select this area. And if we check auto segmentation, if we set the background as static, then it will remain the moment path. It won't consider this as a static area. So let's reset everything. And in this clip, I want to move some portion of my character, not a whole character. So what I'm going to do is I'm going to select my arm of this character. Let's draw this arm to his face, select the area number two, and I'm going to draw the arm of this character. Let's draw this as well Let's uncheck let's draw this arm as well. And now I'm going to block this and now we are done. Let's let the background, first of all, uncheck this. Let's draw this. Let's draw his face only. And now I think we turn, we can select the static area, click on the confirm, and let's generate this clip. So the video is now generated. Here we have the result. As you can see, guys, one got is moving to this position, other is moving to this position as we set in the motion brush. Now in this example, I draw a path expect to make me fly in this video, but instead, we got this result. Which is not bad. And in the next example, I draw a line expecting him to make me sit, but we got this result. Now in the next example, I found a solution to this problem. So in this example, I draw this path, and now I am now standing in the video. And in this example, I'm now finally sitting. So now I will show you how motion barrage works. So if we look closely to the path, as you can see, guys, I draw a path between my legs. So the AI is now understanding that I'm asking him to make me sit. And now I will talk about what's the problem with the other generation. So if you look closely, I overdraw my arrow onto the drawing area, that is why we got this result because the AI is now confused. Whereas, as you can see, guys, in this video where I am sitting, I draw on the static area. And now in this example, we got this result This robot is not raising its hand. It's not raising its punch. So in the next example, what I did, I just draw this arrow and we got this result. So you have to draw motion. So now in this example, we got this result. And I changed the arrow a little bit. As you can see, guys, now I change my arrow path a little bit, and we got this result. I have this example as well. As you can see, guys, if we open the motion brush, I draw this character completely, and I draw a path. I want this character to move to this position. And for this character, I only draw the punches and the head of this character, and I set the background as static. If you look closely, this character is not completely set to this static, only its punches and the head is now drawn, not the whole character, and I draw this path where I want to move him. And this is a result of this drawing. And this is another simple example if we open the motion brush, I just draw my car and I just draw a path. Set the background as static. We have another example as we robot fighting. If we open the motion brush, I draw this character, I want to punch this character, and at the same time, I want this character to lean back a little bit. And for this character, I draw this punch. I want this character to punch this character and also lean back a little bit as well. And we got this result. You can also draw multiple areas, and you will get different kind of result. This is another simple example as well. If we open the motion brush, I draw this character to this side, draw the bus and move the bus to this side, set the background as static, and now we got this result. The robot is not moving because the AI may be thinking that I want the bus to crush this robot. So right now, the AI is sensitive as well. We have another robot example as well. If I show you the motion brush, here we have the motion brush. Now here we have the result. As you can see guy, the results are stunning. Now if you use a professional mode, you will get precise result. So in this example, I draw the left arm of this and also the left arm of this robot as well, and I draw this path. Set the background as static, and it generates this result. The result is now precise. Unlike the standard mode, with the same drawing, if I show you with the same drawing, I generate this video with the standard mode, and the character also moving a little bit and background is also moving. But in the professional mode, you can see, guys, it is precisely only moving the part where I drawn. So with this drawing, we got this result with the professional mode. So as you can see, guys, everything is now consistent and with the same drawing. But with the standard mode, we got this result. Now you can also move your edited image. As you can see, guys, I edit myself in the studio, and I draw this simple motion brush. We got this effect. We have another example. If I show you the motion brush, this is the motion brush, and this is the result with professional mode. The professional mode video look weird because it generate what I ask him to. So as you can see, guys, I ask him to move the muzzle flash and his arm as well, and it's just did and it look weird. But with the same drawing with the standard mode, you will get different result. And this is the result with the standard mode. As you can see, guys, it's moved the arm and the muzzle flash as well. The motion brush with a standard mode, it look more realistic and more stunning than the professional mode. But the videos will inconsistent if you rule the standard mode. But with the professional mode, you have to be precise with your brushes because it will only move that portion where you brush the area. So if you are wondering how I created the board punching videos, this is the prompt and these are the result using the Microsoft designer. Now in this example, I only draw the flash of the gun and we got this result. Let me show you the motion brush. As you can see, guys, everything is not set to static except the face and the body of this person. So my point is you have to be precise while using the professional mode if you use the motion brush, if you want to get the motion right. And this is how you can use the motion brush. 23. Lip Sync Videos With AI: Welcome back. So now we have the final feature, which is called lip sync. If there is a pace in your clip Ino generated video, you can lip sync that person. You have to just select your video. You have to click on the lip sync. It will first identify your video. Then we have this interface. As you can see guide, there is so many voices. You can select any. Hey there. You can play and you can do. You can listen to that voice. You can select any gender, like male, young man. Hey there. Beautiful. Once you select your voice, you can type anything that you want him to say. Once you type your speech, you can also click here to preview your words. Hey, there. I am from the past. My name is Alex. You can also change the speed of your voice. Hey, there. I am from the past. Let's change this to one X, which is the normal speed. And now you can click on the lip sync and it will take your five credits. Now you can also upload your audio as well in order to lip sync. So right now, I'm going to click here, Lip sync and go to the upload. Local dumbing. We can upload a file. And as you can see, guys, if your recording is more than 5 seconds, then you have to corp your audio. Hey, there. My name is Faizan and I am from the past. Hey, there, mine. So you have to select a portion from your audio because the video length is five second, so we have to select the audio under five second. Now I'm going to confirm my crop, and now let's lip sync this clip as well. It doesn't matter if you use image to video or text to video generation. If there is a phase, then you can lip sync that video. So for example, I'm going to select this robot, and now let's click on the lip sync. So first, it is now identifying if there is a phase to lip sync. Now if it fail to identify a pace in your clip, then you will get this error. Let's identify this video, and now as you can see, guys, now we have this interface. Now if you want a person to speak very slowly, then you can set the speech rate to 0.8, which is the lowest. And before you type anything, you can preview that voice. Let's walk to this side and maybe I put my hands in my pocket. As you can see, guys, the lyric is now speaking very slowly, and if I want him to speak fast, you have to increase the speech rate. Let's walk to this side. And maybe I put my hands in my pocket. And there is a duration limit up here on the right side of the voice. Let's walk to this side. So this voice is now 4.5 second. So maybe we have to change the speech rate to 0.9. Let's walk to this side. Now we have 5.3 second voice. Now we can preview this voice. Let's walk to this side. Now, if you want a person to speak very fat, then just increase the speech rate. Let's walk to this side, and maybe I put my hands in my pocket. For me, I'm going to set my speech rate of 0.9 and click on the lip sync. So the lip sync video is now generated. Here we have the result. Hey there. I am from the past. My name is Alex. Now, as you can see, guys, the results are now pretty amazing. Now, if you have a perspective video like this is the front camera, but if you have a side camera, then the lip sync is also very stunning. Hey, there, my name is Faizan and I am from the past. As you can see, guys, my perspective is now changing, but it got my mouth movement. My name is Fazan and I am from the past. We have this example, as well. Let's walk to this side and maybe I put my hands in my pocket. I think my mouth look way too big than it should be, but it's okay. It's not that bad, and this is the final video that we generated with lips. Oh, I'm getting fat. My lungs, my lungs. Now you can also redub your video if you want. You just have to click here redub you can upload your audio or you can use text to speech in order to lip sync your clip. So this is how you can use the lip sync in order to lip sync your videos using cling AI. 24. Extend AI Video Generation: Welcome back. Now, if you generate any video, you can also extend that video. So in this example, I'm going to select my burger right here. I'm going to enhance this burger. Now, the burger is enhanced. Let's bring this to life. Now I'm going to select professional mode and let's select cling 1.0 because if you use 1.5, then you can't really extend your clip. There is no option. If you are using 1.5, you can't really extend your clip. So the video is generated. Here we have the result. Now, if you look closely, we have an option called extend, and we have two option in extend. One is auto Extend. The other one is customized extend. Now, if you are using the standard version, then you can also use this as well. And it will take ten credits if you use the standard mode. And if you use the professional mode, it will again take 35 credit. So let's extend automatically. Now I'm going to select my generated video again and this time, I'm going to select customize extend. Now I'm going to type catchup pouring on top of the Burker. Now the video is extending. So we have the result. As you can see, guys, the video is now extend to nine second. And this is auto extend result. Now we have the customer dessert. And after 5 seconds, as you can see, guys, we have this catcher pouring on top of the burger. And just like that, you can extend your clip. You can type anything that you want in your video. Now I have another examples as well. So in this clip, we have this prompt. I generated this video using text to video And by the way, this is the standard mode, not the professional mode. And this is also the standard mode. And we can also hoard the mouse to the extension history. And as you can see, guys, I have typed a hand powering milk. So we have this result. We don't have any milk, but you can try again again to get your desired result. Sometime it does not extend what you ask for him. So in this clip, I type dark cloud covering sky, and as you can see, guys, there is nothing happen after 5 seconds. It's just a normal Zoom in video. There is nothing happening in the sky. This is another example where I type seen move to a car parking. And if I play this video, as you can see, guys, after five second, we have a car behind the character. This is by the way, the auto extend. We have just a normal Zoom out clip. And I type a futuristic bike is in front of the arrector. And after five second, you can see we have a bike. This is also a standard mode video. I typed car crashing and after five second, you can see we have this smoke coming from the car. Sometimes it does not generate what you ask for him to generate. So in this example, I ask him to take the letters off from the burger, and it just generated this video. And in another example, a hand is adding extra patty in the burger, and we have this hand is adjusting the burger, not adding the patty. And in this example, I type Imodi is turning red and getting angry. So after 5 seconds, we have this result. He's getting angry, but he's not running red. Not only you can extend your video to up to 10 seconds, you can extend to as much as you want. Like in this example, in my extension history, this is the original prompt that we generated this video. And after I add plane passing by, then I add aeroplane is flying in a distance. Then in the last prompt, I add sky is getting darker. Rain is about to fall. So this is the whole video. It does not add any aeroplane at all, but the sky is getting darker, which I notice. Here we have after 15 second, the sky is getting darker. So this is how you can extend your clip using King. 25. kling 1.0 vs kling 1.5 Pro Mode: Welcome back. So now we have all the videos generated with Kling AI with different mode. And in this lecture, we will compare 1.0 per factional mode versus 1.5 per factional mode. So right here, we have the setting for the following generated video. First is 1.0 per factional mode result. And now we have 1.5 professional mode result. As you can see, the 1.5 deliver what I ask him to, while the 1.0 deliver just a video of a car crashed. We have another example. Here we have the prompt, here we have the settings, and this is 1.0 professional mode result. And with the same settings, this is 1.5 pro mode. And we have another example. With this setting, we have 1.0 promde result, and now we have 1.5 promote. So there is a huge difference between 1.0 promode versus 1.5 promote. Let's see another example. With these settings, we have 1.0 promo result. And with the exact same setting with 1.5, we got this result. So did you notice something? Let's see another example as well. With this setting, this is 1.0 promo result, and with the exact same setting, we have 1.5 promote result. With this setting, this is 1.0 promote result. And with the exact same setting, we got 1.5 promote result. If done with an animated character video, what it would we look like? With this setting, we got 1.0 promo result. As you can see, guys, we got our face animation, and with the exact same setting, we have 1.5 promote. Now let's talk about if you bring your pictures to life, what it would look like. So this is the result of 1.0 promote, and this is the result of 1.5 promote. So if you're talking about bringing your pictures to life, there isn't any much difference between 1.0 versus 1.5 promote. So 1.5 promde is more consistent than 1.0 promote. 1.0 generates stunning videos, but there is some distortion in the videos that can be noticed. But there is not huge distortion. If you use the standard mode, you will get huge distortion or infringement. But if you use 1.0 promo, you will not get a huge distortion or infringement. But if you consider 1.5 promode, it would be better to use it because there isn't much distortion or infringement that I notice. While comparing these two, I notice 1.5 is more consistent and deliver the best than 1.0 promo. 1.0 promote, in some scenarios, perform better than 1.5. So this is the conclusion of whether you want to use 1.5 or 1.0 promote. 26. Kling 1.5 VS Kling 1.0 Standard Mode: Welcome back, everyone. So the new update of Kling AI allow us to use Kling 1.5 in the standard mode. So before this update, we can only use the professional mode. Now we can use the standard mode, which is a good thing. And the standard mode will take 20 credits. First of all, I'm going to select image to video model and let's select an image. Now I'm going to select my image, and now I'm going to type this brown man running in a cyberpunk city wearing a cyberpunk style suit. Now let's select the standard mode, and let's click on Generate. Now, I have generated some few examples, and now we are going to compare Kling 1.1 versus 1.5 standard mode. This is the first example, men running in a cyberpunk setting wearing a cyberpunk style suit. And as you can see, guys, I use Kling 1.0, and I have used the standard mode. And this is the result. And with the same image with the same prompt, this is Kling 1.5. And as you can see, guy, the result is improved a lot. I did not even use the professional mode for this one, as you can see right here, I have used the stand in mode. Now let's look another example. Giant slime far from the ceiling and covered the man's body. Now I use Kling 1.5 professional mode for this one, and this is the result of the professional mode, Kling 1.5. And with the same image with the same prompt, I just use the stand in mode of Kling 1.5, and here is the result. So you can use Kling 1.5 standard mode, and you will get some pretty amazing results. Here we have another example, man running from the tsunami arriving from the background. I use Kling 1.0 model and I use standard mode. And here we have the result. And with the same image and with the same prompt, this is the result of Kling 1.5. And as you can see, guys, the results are pretty consistent. Now here we have another example of ing 1.0 standard mode, man running from the big explosion happening in the background camera following him. With the same image with the same prompt, this is the result of Kling 1.5 standard mode. Y. Now this is the final example. Everyone starts fighting, punching each other. I use Kling 1.0 and use the standard mode for this one, and here we have the result. And with Kling 1.5, here we have this result. So now you can use Kling 1.5 standard mode and you will get some pretty amazing results if you don't have a professional account of Kling AI. 27. Start and End Frame Kling 1.5: Welcome back. So in the new update, you can also use start and end frame by using 1.5 cling AI model. Now, I generate this clip by using these two frame, this is the start frame, and this is the end frame. By the way, you can also swap the end and the start frame if you want. And I use the professional mode for this generation, and this is the result. And as you can see, guys, pretty much all the movement of my body is inconsistent. So let's change the model to 1.5, and let's see if the video will stay consistent or inconsistent. We have another example of start and frame. By using the standard mode, this is the result. Now let's change more to professional and let's change the model to 1.5. We have this example as well. By using 1.0 professional mode, this is the start frame, and this is the end frame. And we got this result. Now let's see if we got some consistent result by using 1.5. So let's change more to 1.5, and let's click concenrate. So the video are now generated. Here we have the final result. So the video is pretty consistent. But there is inconsistency on my face that I'm not seeing. But it got the motion right. It is improved than the professional mode of 1.0. Now let's check out the second example. And I think this result is pretty amazing than the standing mode of 1.0. Now if you use inconsistent images, if you use the first frame and second frame is completely different than first frame, then you will go some weird result. And this is the first example of that. In both images, the car is stuck in wall. That is why we got this result. Even if you use 1.5, you will go some weird result. Now, this is the second example. If you use completely different images, the start frame is different from the end frame. That is why we got this result. And as you can see, I am using 115 professional mode. Now in this example, I use a Ferrari. If you want to generate consistent images, then first of all, go to designer dotmicrosoft.com. And for the Ferrari, I just type a wide view of a Ferrari car. And by typing this word, you have to just enhance prompt, click on this enhanced prompt, and it will write the prompt for you. Now you can copy or you can select the size or generate the images from designer dotmicrosoft.com. But the images of Designer Microsoft is look like cartoonish. They don't look realistic. For generating realistic images, go to labs dot Google and go to image effects. As you can see, guys, I have select image effects. Let's go to the My library. And as you can see guy, the images of labs dot Google is pretty realistic. Now, by clicking here, you can go to Image effet you can paste your prompt and you can click on this generate. While it is generating the images, let's go to the designer, and as you can see, guys, the designer images look cartoonish. Whereas if you go to the labs dot Google, we got realistic results. Let's download this image. And if you scroll up into your prompt, as you can see guy, this is a wide angle. We can also change this to close up. We can also change the color of the Ferrari. We can change the brand name. You can change the colored keyword if you want. Now we can click on Generate. Before we generate a wide angle, now we can generate a close up of a Ferrari. And by doing this, as you can see, guys, you will got some consistent images. Now let's click on this download button. And now if we open the images, these two images look consistent. They look the same. So after we downloaded the images, I select the close up at the start frame and select the white shot at the end frame, and we got this result. And by the way, this is generated using the standard mode of 1.0. And with the same images, we got this result. As you can see, the result is consistent. If you want the camera to move, you can also swap the frame. And in this result, I just swap the frame. I select the white shot at the start frame and change the close up shot at the end frame. And this is the result that we got. And these two images also generated by using the laps dot Google, and we got this result. So there is another method if you want to generate consistent images. First of all, go to Chat GBD and ask him to write an image generation prompt, a water bottle fall from the sky in Paris. And we got this prompt and now ask him to write a prompt of the second scene. Now I just ask him to keep these two prompt consistent and give me these two prompt. I copy and paste this prompt into labs dot Google. As you can see, these are the first prompt result, and these are the second prompt result. The bottle is falling from the sky, and in the second prompt it landed on the ground. And here we have the result. Now I have used 1.0 tenon mode, but you can change your model to 1.5 and select professional mode, and then you will got some professional, consistent results. Now, you can also ask JA GPT to generate images for you. As you can see, guys, I ask him to write six prompt, keep the consistency in six, a time lapse of a plant growing to a full grown tree. And these are six prompts, but I ask him to generate these images and generate these images. And then I ask him can you make them realistic? And we got these results? Then I asked him to make them ultra realistic and we got these results. Then I put these two images and check out the result. So by using the consistent images, you will get some professional result. Now here is another example if you swap the frame of the form, we got this fade out transition in between these two frame. And if you use this frame at the start and this frame at the end and use standard 1.0, we got this result, which is better than 1.5 result, but 1.5 result is consistent. So I hope you get the idea how you can generate the start and end frame videos professionally. Just keep images consistent. Now, if you have two consistent images, then you can generate some pretty amazing result. Like I have these two images which look identical to each other. In the first frame, we have a seed, and in the end frame, we have a small plant or a tree. Now, I use link 1.0 standard mode, and we got this result. A seed becoming a plant or a small tree. Now I just change my model to 1.5 professional mode, and I got this weird result. The seed is not transforming. We got just two clips put together in a video. So if you are getting this kind of result, then try to type a prompt. So I just use these two images again and type a prompt, seed transforming into a small tree. And then I got this amazing result. 28. Motion Brush of Kling 1.5: Welcome back. So the new update of Kling AI allows us to use Kling 1.5 with the draw Brush feature. Now you can use draw Brush feature with Kling 1.5. So now we have this example. As you can see, guys, I have generated this by using draw Brush feature. And this is my draw Brush path, and this is a result of Kling 1.0 standard mode. Now with the same image with the same drawn path, I'm going to use the professional mode, and I'm going to select cling 1.5 and let's generate. Now we have another example. And if I go to the path, as you can see, guys, this is my path. Now I'm going to use the professional mode and let's switch the model to 1.5, and let's click on Generate. Now we have the same image this time. But I only drawn the flash of the gun, as you can see. Now I'm going to select professional mode and let's select clean 1.5, and let's click on Generate. This is another example of motion brush with professional mode and using and this is another example of motion brush by using clean 1.0 model and professional mode. And if we go to the path, this is the path that I drawn, and here we have the result. Now, let's change the model with the same motion brush path drawn. Let's click on generate. And we have this example as well. With this image, if we go to the motion brush path, this is the path that I drawn. With a professional mode of 1.0, this is the result. If you watch my previous lecture on motion brush, I try really hard to make me sit. And at the end, I got this result after some few tries. So now let's try the same motion brush path, but with Kling 1.5, and let's use the professional mode to get access to motion brush of 1.5. And now this is the result of Kling 1.5 professional mode. And as you can see, guys, everything is now stay consistent. And this is the second example of 1.5 professional mode motion brush. And as you can see guide, the result is pretty amazing. And this is another example where I just brush the flash of the gun. And this one is pretty wild. And this is another example of tour boot. And as you can see, guys, the motion of their body is consistent. There is no infringement in video in any frame. It's kind of look like a real three D animation. Even this example, whereas in the professional mode of 1.0, we got a weird hand movement, but in 1.5, I noticing the hand movement is now fixed. So 1.5 is actually improved in motion brush, and this is the result of me trying to seat myself using the AI And I think this result is pretty amazing. And here we have some other amazing examples. So I use this image, and if we go to the motion brush, I draw this path. First, I draw the pedal and track this pedal to this side. Then I draw the whole water and track the water to that side. Then I remove this side of the water because if we look at the image, we have these ripples in the water. So I remove this part. So I remove this path by using the arrays tool, and then I select the area three, draw this path and track this pad to that side. And final, I select my host subject and track my subject to upper side. Now at the final, I draw this board as a static area. And I use standard mode for this motion brush generation and use 1.0. And if I play my video, this is 1.0 result. Even 1.0 is looking amazing. Now by using 1.5 with same image, same motion brush path, we got this consistent result. Isn't it amazing? Now you can do one more thing. If you notice we got a problem. The boat is not moving at all. So what you can do as well is you can select the As tool, enable auto segmentation and erase the static area. Because we draw the boat as a static area, that is why the boat isn't moving. Now we can select the brush tool. We can select area five. We can draw the whole boat. And we can track the boat to this site. Now let's confirm And for this, I'm going to use my standard mode and select 1.0. While the video is generating, let's view some other example. So I use this image, and if we go to the motion brush path, now in the motion brush path, I draw my car, and I want to move my car through this side. And I draw the whole area as a static idea. With 1.0 standard mode, this is the final result. As you can see, the car is going this way. So I fix my area in the next generation. If we go to the motion brush, I draw my line to this side. In previous I draw my line like this. Now I have drawn to that side. And here we have the result of 1.0 standard mode. And it looks like the car is kind of drifting. And with 1.0 professional mode, with this updated path, here we have the result. And with the same car with the same image, same path. But this time, I use 1.5 professional mode, and this is the final result. Now we have another example as well. I use this image, and if we go to the motion brush, I draw my car and I draw this path. I want to move this car to this side, and I draw the whole ocean and I move the ocean to this side and draw the whole area as a static area. And this is the result. This is not what I expected. The car is going backward instead of going forward. So in my next generation, I fix this problem. I draw this path to this side, I draw the arrow to this side. In previous generation, I draw the arrow like this side. So by drawing the arrow to this side on the ground, I got this result. We fix this problem. So if you got these kind of problem, then you can easily fix them by conducting your area. And by using 1.5 professional mode, this is the final result. As you can see, the video is consistent. There is no infringement. Everything look realistic. And we got this example where we have a tiger. And if you go to the motion brush, I draw the tiger and I point the arrow into the water and I draw the whole water and point the arrow to this side. I want the water to move to this side, and I want my tiger to go inside the water. And this is what we got by using 1.0 Now, if you go to the motion brush, I draw this rock as attic area. But in my next generation, if we go to the motion brush, I remove this as attic area. And I also correct my arrow positioning of the tiger, and with 1.5 professional mode, this is the final result. This is the result that we want. The boat is kind of moving. Not the boat, but we got a camera moment in the video. Now let's use a professional mode, and let's use 1.5 and let's generate. So if you don't want to lose any additional credits, I have a tip for you. If you want to use motion brush, then use 1.0 tendon mode. By doing this, if you got something wrong, then you can fix your problem in your next generation. And if the problem is fixed, then you can use 1.5 professional mode, and you will got your generation, and the video will stay consistent. This method allow you to save your credits. If you use 1.5 professional mode in your first generation, then you will lose 25 additional credits. So this is my way to save the credits. You can use the same method to save your credits as well. So 1.5 professional mode video is now generated, and here we have the result. So this is how you can use Motion brush tool by using 1.5 professional mode. 29. Camera Movement With Kling 1.5: Welcome back. So in the previous version of ling AI, we can't really use the camera moment if you switch image to video inkling 1.5 or inkling 1.0 as well. But now we can use the camera moment in cling 1.5 by using the professional mode. So to generate images, I use designer domicrosoft.com, copy this prompt. Then I paste the prompt in labs dot Google. And select image eft. Now you can also change the highlighted keyword. If I go to my library, as you can see, these are my original prompt results, and I select their highlighted keyword and change some few keywords, and then I got these results. If you go to the Microsoft designer, I select this image, you can click here, edit entire prom, you can copy this prompt, and I paste that prompt in labs do Google, and I got these results. Go to cling AI, select your model to 1.5. Go to image to video. Now first, I'm going to select the mobile image. You can scroll down and you can select professional mode. And as soon as you select your professional mode, as you can see guys, we got camera moment. You can select your camera moment from this drop down menu if you scroll down even more, as you can see guy, these are my camera moment. So if you select this, you can change the slider to get your camera moment. If you change this to left side to minus ten, you will get left horizontal camera moment. If you select vertical, you will got the vertical camera moment. We got the Zoom we got the pen, and we got tilt and roll. So I'm going to select this pen minus ten. Let's change the two minus five. Let's not go to extreme. Maybe let's go to extreme. Now let's click content rate. Now let's select this picture and let's download this image. And for this image, I'm going to select tilt and let's change this to minus ten. Or maybe we can select this image. Let's download this image, and let's scroll down, and I'm going to change my tilt to ten. Now let's click on generate. I also generate this image, and this is the prompt of this image. Now for this image, I'm going to change the pan to ten. While this video is generating, I'm going to show you some few examples of camera moment by using 1.5 image to video. So in this example, I select this image and I change my camera moment to vertical ten, and I got this result. Now in the second image, I change the camera moment to pan minus ten, and we got this result. And for this image, I use Zoom camera moment, value ten, and this is the result. As you can see, guys, 1.5 is pretty consistent. And for this image, I use my camera moment pan, value ten. The camera is panning toward the sky. And if you notice there is no inconsistency in the video, the video is pretty much consistent. And in this example, I use role value ten. This is the extreme camera moment. And still we got some infringement right here, but not too much. Now, by the way, you can also mix these types of camera moment videos together. For example, I can mix this video with this video, let's download these two videos. Now I'm going to show you how we can mix these types of camera videos together. First of all, open the cap cut. This is the best beginner video editor software right here in the market. Now, create a new project. Click here to create a new project. Now select these two videos and import them right here in the media page. Now I can import this video first, and let's import this video beside the first clip. Now we can press space to preview this clip. So the camera is panning to this side, so we can also reverse this clip as well. But I think instead of reversing the clip, let's place this clip right here and we can reverse this clip right here. Click here. First select your clip and click here to reverse your video clip. And now, as you can see, we can put this video right here. We can also select this clip, go to the mask, select the horizontal mask. We can invert this mask by clicking here, and we can select this arrow and we can move this upside to make this feather. Now we can drag this line to this side and change the feather. Two 0%. Now add a keyframe right here, and we can also add the keyframe in position. Now go few frame ahead, change this to this side and change the feather as well. And now, if you check out the video, we got this amazing result. Let's crop this select this clip, go to the animation, go to the out animation, and we can add a fade out animation. And if I play this clip, as you can see, guys, this is the final result. This is how you can use your camera moment clips of Kling AI and merge them together like this. Now, these are some other example of camera moment by using Kling 1.5 professional mode. This is the pan camera moment value minus ten, and this is the result. Now I highly recommend don't use to extreme value of your camera moment. Just say like here, if you want to get some realistic result, two minus three or two plus three, if you want to get some amazing result. Now we have the second example where I use extreme tilt position ten, and this is the result. Now, this is the final example. I use pan plus ten, and this is also a extreme camera moment. But for this shot, it works pretty well. So this is how you can use the camera moment by using 1.5 professional mode. 30. Kling 1.6 vs 1.5 vs 1.0 Text-to-Video: Welcome back. So Kling AI just launched their new model, which is Kling 1.6, and it is set to improve than their previous version. Like Kling 1.5, you can also use standard mode with Kling 1.6. But currently, we can't really use motion brush, and we can't really use camera moment with Kling 1.6. So I have this example. This is a prompt, and I use Kling 1.0 professional mode, and this is the result. Now with the same prompt, I use cling 1.5, and this is the result of 1.5 professional mode. Now let's go to the edit and let's generate this with a new version of 1.6. Now we can click on Generate. So the video is now generated, and here we have the result. Let's view the result. By far, 1.6 is pretty amazing. It is far more better. It is far more superior than 1.5 and 1.0 because you can't really find any frame infringement or you can't really notice any frame infringement in your generated clips. Now, I have some more example that I want to share with you guys. This is the prompt, and I use 1.6 promote for this generation, and here we have the result. As you notice, there is zero frame infringement in the generated clips. And as you can see, guys, the 1.6 is more clear and there is no inframe infringement in 1.6. In 1.0 and 1.5, we have a little bit frame infringement, but in 1.6, there is none frame infringement in the generated result. Now, here we have another example. This is the prompt, and I use also the pro mode for this generation. And the result look realistic. There is a little bit frame infringement right here, but everything looks great. And this is 1.0 and 1.5 result. Now, before you get excited and you start use 1.6, I want to share with you guys this secret. The 1.6 is four realistic clips. If you use unrealistic prompt, you won't get much more better result than 1.5 or 1.0. Let me show you an example. The same prompt, we have both 1.0 and 1.5 result. And now with the same prompt, this is 1.6 pro mode result. There is no frame infringement, but I did not like the style of 1.6. 1.5 is far more better if you want to generate animated clips, but 1.6 is for realistic clips. So with this prom, this is the result of 1.5 pro mode. And with the same prompt, this is the result of 1.6 pro mode. Now, all of these videos generated using 1.6 professional mode, but is 1.6 standard mode also good for video generation? I'm going to show you some few examples, and I'm going to compare 1.0, 1.5 and 1.6 standard mode because all of these videos are generated using the professional mode. Now let's see if standard mode also good enough for video generation or not. So this is the prompt, and this is the result of 1.0 standard mode. Now with the same prompt, this is 1.5 standard mode result. And 1.5 is also good for video generation. And now we have the final 1.6 standard mode. So as I told you earlier, 1.6 is far more realistic than the other, because 1.5 video looks great, but it is unrealistic for me. So 1.6 is for realistic generation, but if you want to generate in between realism and unrealism, then use 1.5. Now this is another example. This is the prompt, and this is the result of 1.5. And here we have the result of 1.6 standard mode. And for this example, I think 1.6 is a clear winner because in 1.6 generation, you can clearly see a little bit camera moment and everything looks crispy clear. Now with this prompt, this is a result of 1.0. And this is 1.5 standard mode result. And now this is the result of 1.6 with the same prompt. And for this example, 1.6 is also the winner for this generation because in the background, the building looks also really great and can also look great. Now we have another example. This is the prompt, and this is the result of 1.0 standard mode. With the same prompt, this is the result of 1.5 standard mode. Now finally, we have 1.6 standard mode result. And as you can see, guys, 1.6 is great. It looks like we are shooting this sphere using the mobile phone. And as I told you earlier, 1.5 is in between realism and unrealism. Now we have second last example. This is the prompt, and first of all, this is the result of 1.0 standard mode. And as we expected, everything looks crap. Now let's move on to 1.5. With the same prompt, 1.5 manage to generate this masterpiece. We have a little bit infringement, but everything looks great. Now finally, we have 1.6 result. And this is unexpected result for me. Because I believe 1.5 looks great in this example. Because in 1.5, you can see clearly in the background, we have some audience, but in 1.6, there is no audience. And as I told you earlier, if you want to generate realistic video, then use 1.6. But if you don't want to generate realistic video, but also you want to generate some of the realism in the video, then use 1.5. Now, if you try to regenerate this prompt, this is the result of 1.5. And surprisingly, 1.6 now perform a little bit better. And if you notice when the robot start to work toward the red robot, the ground shake and with the ground shaking camera also shakes a little bit. And now we have the audience in the background. For this generation, I think it's a tie for 1.6 and 1.5. Now we have the final example. Heat is the prompt, and with this prompt, this is 1.0 standard mode. And with the same prompt, this is the result of 1.5. And now, finally, we have 1.6. And as you notice, guys, even with this standard mode, 1.6 is stand out. Now, I accidentally regenerate this prompt using 1.5, and this is the regenerated result. And 1.5 is look better than before in this regenerated clip. So for this donation, 1.6 is also the winner. So by far, what I have used 1.6 is far more better if you want to generate realistic clips. 31. Kling 1.6 vs Kling 1.5 Pro Mode: Welcome back. Now let's talk about 1.6 image to video generation. Is 1.6 improve in image to video generation than 1.5? Let me show you some few examples. First of all, in this example, we have this picture, and I write a prompt a knife cutting apple into two pieces. And as you can see, I use 1.5 model, and I use professional mode for this generation. Now even the professional mode, as you can see, guys, it's look unrealistic. Now with the same image with the same prompt, I just use 1.6 professional mode, and this is the result of 1.6. And after seeing this generation, 1.6 is far more better than the other tools that are available currently in the market because the motion is great and Apple slice slide looks realistic. Now let's see some other example as well. Now in this example, I use my picture and I type a prom. Room is getting darker very fast, and person head is on fire lighting the room after. So I want the room to get dark and I want my face to set on fire. And as you can see, I use 1.5 professional mode, and this is the result. Look great, but it's not looking realistic. Now let's take out the result of 1.6 with the same prompt and with the same image. I use the professional mode for 1.6. Even the video is unrealistic, but visually, 1.6 is far more better. Now we have another example. And with this picture, I type Man running away from the robot that following him. I use 1.5 professional mode and I got this weird result. And after I see this result, I thought, why not regenerate this clip using 1.6 professional mode? 1.6 is going to be better, right? If I play my clip, as you can see, the person is running, everything look realistic, but there is no robot in the clip. I ask him to add a robot, but there is no robot in the clip. So this is a little bit disappointment. So I want to fix this problem. So I thought how to fix this problem. So in this example, I just change the prom. Person running towards camera, our giant robot is chasing him. I use a different image because in the previous image, you can't really see my face. And I use 1.6 professional mode, and now we have improved result. Person is running, and robot is also chasing him. And as you can see guy, the result is looking really great. And I satisfied with this result. So if you got any problem with your generation, just change the prompt. You will get the result that you want to. Now, if you don't get the result that you want, like in this example, with the same image with the same prom, I got this result. But I regenerate this result, and I got this result, which is improved in quality and in motion as well. If you don't get the result that you want, even if you type a very precise prompt, you got a high quality image of yourself or anybody, try to regenerate thickly because it helps you to improve in quality or in motion as well. Now we have another example. With this image, I type this colorful balloon falling from the ceiling. Some of them fall on the person light illuminating the room with color balloon. And this is the result that I got this is by the way, 1.5 professional mode, and everything looks great. And now we have 1.6 with the same pro with the same image, and also same mode which is professional mode. Let's see the result. 1.6 is far more realistic than 1.5. But in 1.6, we got some detail. If you notice in the LCD, as you can see, guys, as soon as balloon falling from the ceiling, we have the reflection in this LCD. And we also got realistic balloon, but they appear for the shorter period of time. 1.6 is a little bit better than 1.5. If you compare the result, 1.6 is a little bit better, not too much, but a little bit better. But if you talk about the emotion, 1.6, just destroy 1.5 because 1.6 motion looks realistic then 1.5. If you check out 1.5 emotion, they just a feeling that this is not real. But in 1.6, everything looks great. But I kind of like the both results. So if you want to use 1.5, then you can definitely use that. Now, 1.6 is also better in terms of expression. So with this image, as you can see, I type a prompt, a person standing with a neutral expression, transitioning to a sad expression then to a happy expression and then and finally smiling. And this is the result. Now, after that result, I regenerate the result with the same mode with the same model, which is 1.6. Here we have regenerate the result, and I kind of like this result than the previous. So if you don't get the result that you are expecting, then try to regenerate. We have another example of 1.6 professional mode. With that same image, I type man running in a cyberpunk city wearing a cyberpunk style suit, and this is the result that I got. I if you try something unrealistic, you might get some few infringement in your generated clips. But in terms of visual, I think 1.6 is looking great. And now this is the final example of 1.6 professional mode. I use this image and I type card transforming into a white car, and this is a result of 1.6 professional mode. Now, in terms of video quality and in terms of realism, 1.6 try to be realistic as much as possible. 32. Kling 1.6 vs Kling 1.5 Standard Mode: Welcome back. So 1.6 professional mode is better than other AI model. But is 1.6 tener mode also good for video generation? Let's find out with some few examples. Right here, we have this image and we have that same problem that we use in the professional mode. Now, in terms of expression, 1.5 is not really good at it. Now with the same image with the same prompt with the same mode, which is the standard mode, but I use 1.6, and even in the standard mode, 1.6 is better than 1.5. Now we have another example. With this image, with this prompt person running towards camera, a giant robot is chasing him. I use 1.5 standard mode, and this is the result. For this donation, I think 1.5 did a really good job. But the robot is not chasing me. It's chasing these people. And now with the same image with the same prom, but with 1.6 model with standard mode, this is the result. And I think 1.6 did a really great job. It beat 1.5 in this generation as well. We have another example. Image with this prom, everyone starts fighting, punching each other. I use 1.5 standard mode, and this is the result. Now with the same image with the same prom, just use 1.6 for this one, and I also use standard mode for this. And this is the result. I look a little bit better than 1.5 because there is less infringement than 1.5. Let's take out some other example as well. Now, do you remember this guy? I use this moster image for the comparison of 1.0 and 1.5 professional mode. But they both fail to express the expression of this moster. Now with the 1.6 standard mode, 1.6 did a really good job at expressing the expression of this moster. Now let's see some other example. With this image, this is 1.5 standard mode. With this prompt, a dint slime fall from the ceiling and cover the man's body. Now with the same image with the same prompt, with 1.6 standard mode, this is the result. Now, I think 1.5 in this generation did a really good job, then 1.6. Now we have another example with that same image with this prom man running in a Cyberpunk city wearing a cyberpunk style suit, with 1.5 standard mode, this is the result. And I'm not going to lie. I kind of like the result that I got from 1.5, even with the standard mode. But my question is, it did not generate any cyberpungity. We only got a futuristic cube or something, but we did not get any cyberpungity. Now with the same image with the same prompt, 1.6 standard mode, this is the result that we got. And even with the standard mode, 1.6 beat 1.5 in terms of visual and in terms of video quality. Now with that same image with this pro man running from big explosion happening in the background camera following him. With 1.5 standard mode, this is the result. And with the same image with the same prompt, 1.6 tender mode, this is the result. 1.6 perform better in terms of visual and in terms of expressing the expression of the people. As you can see, this person notice the explosion and runs towards the explosion in terms of save people or something like that. I think 1.6 did a really fantastic job in this generation as well. Now you might ask a question what model you should use? 1.6, 1.5 or 1.0. Now, in terms of realism, if you want to create realistic videos, use 1.6 because 1.6 is far more realistic than 1.5. But if you want to generate clip, that is a mixture of realistic or realism plus unrealistic, then try 1.5 because 1.5 generate in between both. But if you want to try out your prompt, if you want to experiment with the prompt and at the same time you don't want to lose any additional credits, then use 1.0 because 1.0 will generate with ten credits was 1.5 and 1.6 will cost you 20 credits each generation. After you experiment with 1.0 generation, you might get some idea what the 1.5 and 1.6 video would look like. So this is my conclusion. I hope I answer your question. Now, if you have some other question in this course, then feel free to ask me. I'm always available to answer your question. 33. How to Lip Sync With Kling AI: Welcome back. So now we have a new update of lip sync. Now we have a dedicated section for only lip sync. Now what is new in this update is that we can upload the video in order to try this lip sync. Before this update, we can't really upload any video. We have to generate a video in order to lip sync that video. Now if you use some videos that contain multiple phases like in this example, as you can see, we have multiple faces, then the AI will fail to recognize any phase. So my recommendation is that use that video that shows your face clearly and that contain only one phase. Now I'm going to upload my video. Now, once you upload your video, AI will analyze your video and then it will confirm that this video use lip sync feature or not. So you can clearly see my face in this video, so now we can lip sing this video. Now, if you want to switch your video, if you want to change your video, for example, I don't want to use this video. If I have to change my video, then I have to click here in order to re select my video. And now, as you can see, we can select any video that we want. Now there is two way to lip sync your video. You can either use text to speech, which contains several voices or you can upload your own audio in order to lip sync. Now I have typed this line. Now let's go to the mail and let's play this audio. Hey there. My name is Faison. I am using King A. Now some voices can't really use emotion, and now we have this message. Neutral emotion is not available with this voice. So we have several voices that can't really use emotion of neutral. So you have to select a different voice in order to use neutral emotion. Hey there. My name is Faison. I am using King AI. Hey there. My name is Faison. Now sometimes in some voices we have this message, so you can't really use that voice. You have to select a different voice. Hey there. My name is Faizan. I'm using Kling Ai. Now let's use this voice. Now you can also change your emotion as well. Let's select angry emotion. Hey there. My name is Faizan. I'm using King Ai. Now let's turn red. Now we have the lip sync. Let's check out the final output. Hey there. My name is Pasan. I'm using King Ai. Now we are facing some problem. As you can see, the lip sync is off the chart. Because if you look closely in the original video, my mouth is moving too much. That is why the AI struggle to reshape my lips in order to try lip sync. Now in this example, where I did not talk at all, as you can see, guys, my mouth is not moving at all. I generated this video using image to video cling AI. Now I try lip sync onto this video and this is the final output. Guys, there are so many balloons in my room. No, believe me, check this out. Now, as you can see, guys, the lip sync is crazy onto this video. Now we have some other example. So with this video, if I play this clip, the mouth is moving a little bit, not too much. And I try lip sync onto this video, we got this result. Hey, there. My name is Faison, and I'm going to create Pixar Animation. As you can see, guy, the lip sync is crazy. Now, as I told you earlier, do not upload a video that contain two phases. Well, I try that and surprisingly, in AI just select one phase. The reason behind is that this one is talking. That is why link focus onto this phase, and I try lip sync onto this character, and this is the final output. You think you can create anime without any special skills? Well, yes, and I'm going to show you how. You think you can create anime without any special skills? Well, yes, and I'm going to show you how. Now we have another example of anime style animation. This is the original source where the character is talk, but not too much, and this is the final output of lip sync. Alright, an hour of work, and I'll be good to go. This is another example, and this is my original source where the mouth is not moving at all of this character, and this is the lip sync result. What a great day for a ride. Got to keep the momentum going. Now we have another example of bad lip sync. So as I told you earlier, if your mouth is moving too much like in this video and in my previous original clip, then you will get some bad lip sync result. How many times do I have to say the same thing? Are you even listening? So if you want to try lip sync correctly, if you want to nail the lip sync onto your video, then apply lip sync onto that video that contain less face lip movement. So if you have a mouth movement too much in your video, then you will get some bad result of lip sync. So this is how you can use the lip sync, using cling AI. 34. Elements: Welcome back, everyone. Now, if we go to AI videos, we have a new feature called elements. And what it does, you can select multiple images, and it will generate videos with that image reference. So for example, let's take a basic example. Now we have these two images. Now, as you can see, from here, you can delete your image, you can upload another image, and here you can select your element. In my example, I want to select this entire image, but for example, if you want to select only this truck, then you can also select this truck. You can click on Confirm, and then you can generate your videos using these elements. Now I'm going to select my entire image. For my second image, I have selected this Ferrari car. So now we are going to make a video with these two images. So I'm going to type a simple prom like a Ferrari car in a field. So the element feature will only work with 1.6 ling model. If you change your model to 1.5, you can't really generate this element video. If you change this to 1.0, we have that same problem. So now we can only generate these element videos using 1.6 model. So with the professional mode selected, we can generate this video. So the video is now generated, and here we have the result. As you can see, we have this Ferrari car in a field. And if I pause this video, if we open the original image, as you can see in the original image, we don't have any cloud. If we open this image, I want you guys to focus on the cloud pattern. And in the generated clip, as you can see, guys, we have that same cloud pattern right here in the sky. And we have this field in this generated video. So I hope you get the idea. Let's see some other examples to better understand this element feature. So in this example, we have two images. We have this image, a plane flying. This is the reference. This is the element, and we have another element. We have a driving motorbike character vector style. Now I typed this problem man riding a bike and plane passing by. With standard mode of 1.6, we got this result. Have the plane passing by and we have that motorbike rider. Now, to generate these kind of images, you can use Labstt Google image effects. And as you can see, guys, if we go to the Labstt Google, this is a prompt data used to generate this image, and this is a prompt for motor bike rider image. Now we can also add an object and we can do anything with that object. So for example, right here, as you can see, guys, this is my original image. Right here, we have this object, a SWOT, IATI man swinging around a SWOT with standard mode. As you can see, guys, in this generated video, I'm swinging this object. We have another example. This time, it copied the background of this object as well. Now in this example, I have three elements. What we have this image of a car in nowhere. We have this background, and we have this subject. And with this prompt, with 1.6 professional mode, this is what we got. So sometimes it generate these kind of videos. So in my next generation, I typed this prompt and with the standard mode, this is what we got. Now I remove one element from these elements, and in my next generation, this is what I got with these two elements only. So in this example, I only have two elements, a man sitting on a couch, and we have this car, and I type this person sitting on a sofa beside an abandoned car in a desert. And even with the standard mode, as you can see, guys, the result is amazing. So the reason I showed you guys this example is because with some simple prompt, we can also generate some amazing results using the elements. So to fix your problems, you can generate more video. You can tweak your keywords of from here and there. Now we have another example with that same image, and this time we have this different car, and we got this result. Card in the field. Check this out. And this time, if we go to the crop mode, as you can see, guys, I have only selected this portion of this image. This is my element that I have selected. Now in this example, we have two anime characters, and I have typed both fighting with the standard mode, this is what we got. Now, sometimes when you generate videos using element, it will take the reference note this element, exactly. I will take the reference of this element and it will generate a brand new video. So with that same images, I have tied both fighting with each other. With the standard mode, this is what we got. As you can see, it takes the reference of these two characters and generate a whole new video. Now again, with these two elements, if I open my image, this is my element number one, and this is my element number two. When sitting on a chair, a big explosion happening in the background. So this is the prom data used to generate this video. Even with the standing board, the result is pretty convincing. Check this out. This is insane. So another reason why you are not getting the result that you want is because you are using the wrong images. So in this example, as you can see, guys, I have these images, and we have these two kid cycling, and this is a weird cycle that I have generated using Labs Google. This is the problem, and this is the weird result. This is another example, and this is the final example. So make sure you have selected the right images. Now we have this example. In this example, I have used for element, which is the most element that you can use to generate videos using elements feature. So first, we have this person. We have this mic. We have this car, and we have this as a background. Now for my prompt, I have typed Boy holding a microphone, sitting in a car parked in desert, camera slowly zooming in. With the standard mode, this is what we got. Now with a professional mode with the exact same prompts and element, this is the result. Now we have another example of most used element. So we have these four elements, and I have tied camera focusing on a cyclist, then focus on plane and then focus on man's standing. With the standard mode, this is what we got. We have cyclist plane. And right here we have a man, this one. Now for my next generated video, I remove the background, and this is what we got. We got this weird result. So the reason why we are getting this morphing effect in the video is because I am using the standard mode. That is why we have this morphing effect. Now, in my next generated video, I fix my prompt eyewek word here and there, and this time it improved the result a little bit. We have Standing Man plane, and we have the cyclist. So you can create some dynamic scene like this as well with the elements. Now in this example, I also died the background again. And with this prompt with the standard mode, this is what we got. Now for some reason, I have used the professional mode with the exact same prompt and the element, and this is what we got. It just generated this brand new video. It takes the reference from my elements, not generate video with the exact same element, but take the reference from my element. You can also improve your generated video as well. So as you can see here with the exact same prompt with these images with this element, if you go to the negative prompt, I have typed this keyword that I don't want these thing in my generated video, disfigurement, distortion, and morph. We look at the original result, this is my original clip, and I have regenerated this clip by using the negative prompt, and this is what we got. As you can see, the result is improved a lot. We have less distortion, disfigurement, and morphing effect in this video than the original. This is the original video and check out the distortion and disfigurement in this video. And now check out the latest video. Now you will also get a different result if you regenerate your clip. As you can see you guys in this example, we have this background. We have the Ferrati card and we have this prompt. With the professional mode, we got this result. Now with the standard mode, I just remove my ferratic keyword from here, place my car first, then place my background second. This is what we got. Now in this example, I place my car first again and place my background second with the exact same prompt. This is what we got. So the reason why I'm showing you guys this is because if you place your background first, then it will copy the element from your background and place that element to your second image, to your second element. Let's experiment this so that you can better understand what I'm talking about. Let's delete this image. Let's delete this image as well. And this time, I'm going to corb my element. Let's remove this rug let's click on Confirm. And we have that same image. With that same prompt with the stand mode, let's create this video. So the video is now generated. This is a preview, and as I told you earlier, guys, if you place your background first, place your car second, we will get less background than the previous example. This is the previous example, and as you can see, guys, we have the background second. That is why we are getting this whole background element in this video. So my recommendation is if you want your background to be static in the entire video, then place your background at the end of your elements. If you have four elements, place your background right here in the fourth element. If you have two images, you want to have a background, place your element to the second as a background. And if you have two elements, and if you want to place a background, then place your element at the third element slot. Now if we go to the Labs, Google, these are my generated images that I use in the element lecture. This is my prompt for this image. For this background, we have this prompt. For the car, we have this prompt. For the anime background, we have this prompt. For the background of the field, we have this prompt. For the car image, we got this prompt. The anime man, we have this simple prompt. We have this prompt and we have this result. We have this image, and this is the prompt. And for this image, I type this prompt. For the final image, with this prompt, this is what we got. So you can use labs or Google to generate impressive images. So this is how you can use elements feature of ling AI. 35. Multi-Element: Welcome back, everyone. So the link AI New Update launched a new feature called multi elements. Now, it is different than normal elements. If you go to normal elements, in normal elements, you have to upload up to four images in order to generate a video. But in multi element, you can swap something in your video. You can even add something in your video, and also you can delete something in your video. Now, currently, 1.6 Kling model is available for multi elements. If you switch your model to Kling 2.0 master, you can't really swap anything. You have to change your model to 1.6. So if I go to image video, in normal elements, we also have 1.6. We don't have 2.0 master. But if you change your model to 2.0 master, and if you go to multi elements, switch to the appropriate model. So as you can see, we can only use multi elements by using cling 1.6. First of all, let's use the multi elements swab. So you have to upload up to five second video if you want to use multi elements. Now, once you upload your media, it will analyze your video. If your video is more than five second, then you have to cut your video. You have to crop your video, and you have to make a video up to five second. So my high recommendation is make your video under five second. Now we have this interface. Now we have to add a selection. We have to select an object that we want to replace. So I'm going to select this car. So I'm adding this keyframe for precise selection. You can also, if you add some point and if you want to reduce or if you want to delete this point, you can even select dio selection, and we can click on it, and it will delete this point. Now let's add this again, and now let's confirm. Once you confirm, you have to upload a picture that you want to replace this car with. So by using this prompt, I have generated this image. As you can see, we have these images. I'm going to download this one, and now we will replace that car with this one. Once you upload your image, you can select subject. So I'm going to crop my image, and let's click on this confirm button. Now, once you select subject, as you can see, we have swap and in bracket, we have X. So we have to name this thing. So I'm going to type vehicle right because we are swapping one vehicle to another. So swap vehicle from this image for this car. So I'm going to also in the bracket, I'm going to type vehicle. Now we can generate this video. Now, as you can see, it swap the vehicle. Now, if you are thinking that you can flip the vehicle in the image and then you will get a better result. So I'm going to show you some example that I have generated before. In this example, I flip my vehicle of this image and everything stays the same. As you can see, we have that result again. I'm going to show you the reason why we are getting this kind of error. So if you look closely, by using the same prom, I also have this image. So I have downloaded this image, and I use the swab video, and in this image, as you can see, I've selected the whole image. I did not crow my image. And as you can see, it also added the mud that we have in the image, and it swap with the rod. In my final example, I did not flip the vehicle, and I crop my subject in this picture. Now, this is the result. As you can see, now, it's just added a little bit mud not too much like we have in this video. So if you want to swap your object, then you have to select a similar object. As you can see, the car is facing this direction, so that is the reason why I have selected this image. That is the reason why this car does not working with the video that we have. Now, you have to also keep this in mind. If you swap a vehicle with the animal, then you will also get some crappy result. In this example, as you can see, I have swapped my picture with a beer, and by using the same video, and in my prompt, I just change the vehicle to bear. And as you can see, we have this result. So you will get crappy result if you swap vehicle with an animal. So you have to swap the vehicle with some vehicle and you have to swap animal with some animal. Now we have another example. As you can see, we have this clip and we have this image. So if I go to the edit, as you can see, I have selected this code, and I swap this code with the dog that we have in this image. Now to generate the image, I use labstt Google Image ex and I type this prompt. Now the reason why I have white background is because the AI work better with a white background. So it will understand better if you have a white background behind your image. Now, as you can see, I swap the goat with the dog, and this is the result. Now again, I use image effects laps dot Google, and I got this result, modern cyberpunk jacket with white background. So as you can see, we have a white background and we have this cyberpunk jacket. Now I use this video, and I have selected my whole subject like this. I have selected this subject like this, and I ddt multiple keyframe to select this subject. Now I have this jacket with a white background. And in my prompt, I type jacket four person jacket. And as you can see, this is the result that we got, and it perfectly swam the jacket with our cyberpunk jacket. Now, what if you want to swap the jacket plus pent? Now in this example, I have generated cyberpunk jacket plus paint with white background. On the same video, I use this image. If I go to the edit, as you can see, we have cyberpunk jacket and paint, and I just type jacket and paint for person jacket and paint. And as you can see, we have this phenomenal result. You can easily swap anything in your video if you use the picture with a white background because with a white background, the AI better understand and it will easily can track this image onto your subject or object. Now let's talk about how we can use the multi element add feature. The add feature work like if you want to add something in your video. So by using the colors 2.0, by using this prompt, I have generated these images, and I use this image. So I have used my video where I was talking, facing camera, and I use this as a subject. Now, it will automatically write this prompt for you. You have to just type in this bracket. Like I'm going to first of all, refresh this so you can better understand how to use the add element. Now let's go to the multi element. And if you go to the ad, first of all, you have to select a video. As you can see, it automatically write this prompt for you. Now, once it analyze your video, you don't have to select anything like we selected in swab feature of multi element. Now as you can see, once it select your video, you have to upload an image. Now, once we have the image using the context of this reference video, seamlessly add and we have this X under bracket. So we can type Robot is sitting on person shoulder. So you have to type your contacts inside this closing bracket from image. Now, if I go to my video that we have generated before and if I go to the edit, as you can see, inside bracket, I typed robot sitting on man's shoulder, and this is the result that we got. As you can see, it seamlessly added this robot on my shoulder. I have some other example of adviiture of multi element. Now in this example, as you can see, I have multiple images. I have bubble image, and I have this creature image. Now using the context of this reference image, simlessly add and I typed multi bubble in the air from this image. A creator is standing beside man, beside person from this image. And as you can see, I only got the bubble. So I again retype my prompt and if I go to the edit and in this prompt, I just type a creator next to the person. And again, I did not get the creator, I only got the bubble. So I fix my problem. In my next prompt, if I go to the edit, as you can see, we have a dog in white background and I have bubble with black background. So I have tied bubble in the air from this image, dog standing next to the goats from this image. And as you can see, it only edit the dog. The reason why it edits the dog seamlessly is because we have the dog in the white background, as I told you earlier. So in my next example, as you can see, I have generated this image with the white background. If I go to the Labs dot, Google Image Effex, this is the multi bubbles with white background. And as you can see this time it addit both the bubble plus the dog. So as you can see, we have cyborg robot dog realistic and we have this dog image. So I use this Cybergrobot image of the dog, and I type Cyberg dog next to the gods. And as you can see, we have this crazy result. As you can see, it matters the light plus, we have the shadow of the dog, as well. Now let's talk about the final feature of multi element, which is called delete. And this is how it works. First of all, you have to upload a video. After you upload a video, it will analyze your video if your video is under five second. Once it analyze your video, again, we have that interface. We have to select an object or subject that you want to delete from the video. So I'm going to select my subject. You can add multiple points if you want to. Now in the final keyframe, let's add again, multiple points, and let's click on Confirm. Now you have to type that you want to delete from this video. Have selected the subject, so I'm going to type delete person from this reference video. Now I have already generated this video, so I don't have to generate again. Now if I play my clip, as you can see, it deletes the subject precisely. We don't have any fill out error in this area. So by using the delete option, you can easily delete anything from your video. So this is how we can use the multi elements feature of cling AI. 36. Kling 2.0 Master: Welcome back, everyone. We have a brand new video model called Cling 2.0. If you go to the Cling AI website, you can go to the Global, and as you can see, we can go to Video, and right here, you can change your model to Kling 1.6 to cling 2.0 Master. Now, if you want to generate text to Video, you can go to TextVdo. You can type in your prompt. You can also get the help of Deep Seek to type your prompt. You can ask him to enhance your prompt as well. Now we can also use the preset. As you can see, you can use this preset in your prompt like your lens, short type, light and shadow, et cetera. You can also use Image to video with cling 2.0 Master. You can upload your image, type your prompt, or you can just use your image and generate video. Now, let's compare cling 2.0 Master with other model. As you can see, we have this prompt, and Kling 2.0 Master is so damn expensive. It will take 100 credit per video of 5 seconds to generate. This is my prompt, and this is the video that we got. And I'm going to be honest with you. This is the worst video if you compare with other model. With 100 credit, this is what I got, and this is the worst video if you compare this video to other model, as you can see, the other model perform better than Kling 2.0 Master. We have this example. As you can see, we have a lego shark in the ocean. The design of the shark is like the other ordinary shark. Now let's compare this video, this prompt to Kling 1.6. Now I have this prompt, and as you can see, I use Kling 1.6 professional mode, and this is what we got. In this video, I really like the design of the shark, but the Kling AI 2.0 master pushes the video to a realistic approach. Now we have this example, Ferrari car transform into Bugatti. This is what we got with Kling 1.6 professional mode with the same prompt, this is what we got. As you can see, even we have some error in the video, we are getting a video. But if you compare this to Kling 2.0, it just give you this ordinary transition from one car to another. Now we have this example and as you can see, it looks like a real life video recording. I really impress now if you use colors 2.0 master of Kling AI image generation model and you restyle your image, as you can see, I restyle this image into 90s enim style. And if you turn this image into video by using Kling 2.0 Master, as you can see, we have this impressive result. Now you can generate high definition anime style videos like this one. Now I use this prompt and I have generated these images. And I turn this image into video by using clean 2.0 Master, and this is what we got. Again, I really impressed by 2.0 Master because the dynamic moment is crazy. Now I'm going to show you this example. This is the result of 1.6 professional mode, POV of a car, and this is what we have. Not bad. Now let's compare these two Kling 2.0 Master. This is with the same prompt. This is Kling 2.0 Master. And as you can see, it looked like a real life video recording of a GoPro. In most of the cases, Kling 2.0 Master performed incredibly great if you compared the Kling 2.0 master to other model, but in some cases, it just perform far more below than the other models. Like, it just gives you an image sometime in this prompt like I got this image, and I have this simple pan animation for 100 credits, not even a zoom in. So you have to keep this in mind if you want to use Kling 2.0 Master. You don't want to lose your credit, then my high recommendation is use Image to Video model because in to prompt, I use Image to Video model, and I really impressed by image to video of cling 2.0 Master, because as you can see, it just gives you a dynamic moment that gives life to your image.