Learn To Combine Real Footage With AI (ChatGPT + Kling) | Vladislav Sateev | Skillshare

Playback Speed


1.0x


  • 0.5x
  • 0.75x
  • 1x (Normal)
  • 1.25x
  • 1.5x
  • 1.75x
  • 2x

Learn To Combine Real Footage With AI (ChatGPT + Kling)

teacher avatar Vladislav Sateev, Video Editor

Watch this class and thousands more

Get unlimited access to every class
Taught by industry leaders & working professionals
Topics include illustration, design, photography, and more

Watch this class and thousands more

Get unlimited access to every class
Taught by industry leaders & working professionals
Topics include illustration, design, photography, and more

Lessons in This Class

    • 1.

      Welcome! Start here

      1:09

    • 2.

      Define Your AI Video Goal & Niche

      0:28

    • 3.

      The Real + AI Video System: Animate from Last Frame or Image (Kling Workflow)

      1:43

    • 4.

      How to Write Powerful AI Video Prompts with ChatGPT

      2:01

    • 5.

      Choosing the Best AI Tool for Your Video (Kling vs Other AI Options)

      2:11

    • 6.

      Generate or Edit AI Video: Blend Real Footage with AI Scenes

      2:26

    • 7.

      Create Longer AI Videos: Extend Scenes and Combine Multiple Clips

      0:53

    • 8.

      Last Step!

      0:44

  • --
  • Beginner level
  • Intermediate level
  • Advanced level
  • All levels

Community Generated

The level is determined by a majority opinion of students who have reviewed this class. The teacher's recommendation is shown until at least 5 student responses are collected.

76

Students

1

Projects

About This Class

Most AI video looks fake. This AI video class shows you how to blend real footage with AI so your videos feel believable and professional.

You’ll learn how to choose a clear video idea, turn it into a strong prompt with ChatGPT, and pick the right AI tool (including Kling, Sora, Seedance, Higgsfield) for the look you want. Then you’ll generate your video, edit it, and blend real footage with AI.

Everything is taught step by step, with short lessons you can follow in one sitting. You’ll watch me build the workflow from the first decision to the final export, so you always know what to do next. You’ll also learn how to extend scenes and combine clips, so you can create longer videos without losing quality.

By the end, you’ll have a repeatable process you can use in any niche. You’ll finish a complete video and feel confident making the next one even faster.

What you’ll be able to do

  • CLEAR VIDEO DIRECTION — Choose the exact outcome before you generate anything, so every clip has a purpose
  • THE SIMPLE AI SYSTEM — Animate from a last frame, photo or transform the video so your scenes stay consistent and controlled
  • PROMPTS THAT WORK — Build strong prompts quickly, so you get usable results without endless retries
  • RIGHT TOOL, RIGHT LOOK — Pick the best AI for your style and niche instead of wasting time in the wrong app
  • BLEND REAL + AI — Combine real footage with AI scenes so your video feels more believable
  • LONGER VIDEOS MADE EASY — Extend a scene using longer source video or stitch multiple clips into one longer story when you need more time
  • REPEATABLE WORKFLOW — Follow the same steps for any niche, so you can create faster every time

With over 10 years of experience editing videos for YouTube, Instagram, and TikTok, I know what it takes to create content that performs. I’ve managed two of the biggest YouTube channels in their niches, and the videos I’ve edited have generated millions of views across platforms.

The workflow taught in this class isn’t theory—it’s the same system I use every day in Higgsfield to create high-performing content for clients and creators. You’ll be guided through the same tools and structure I use to edit with confidence—whether you’re building your own channel or creating for others.

I’m excited to see what you create.

Meet Your Teacher

Teacher Profile Image

Vladislav Sateev

Video Editor

Top Teacher

Hi there! Welcome to my profile. I'm so glad you're here.

My name is Vlad, and I specialize in helping YouTubers elevate their content through professional video editing.

On Skillshare, I share detailed, step-by-step classes that break down my editing process into easy-to-follow techniques designed for creators of all levels.

If you're looking to create engaging, viral videos that keep your audience hooked, check out the classes below.

I'm excited to help you level up your skills and achieve your goals. Let's create something amazing together!

oVlad

See full profile

Level: All Levels

Class Ratings

Expectations Met?
    Exceeded!
  • 0%
  • Yes
  • 0%
  • Somewhat
  • 0%
  • Not really
  • 0%

Why Join Skillshare?

Take award-winning Skillshare Original Classes

Each class has short lessons, hands-on projects

Your membership supports Skillshare teachers

Learn From Anywhere

Take classes on the go with the Skillshare app. Stream or download to watch on the plane, the subway, or wherever you learn best.

Transcripts

1. Welcome! Start here: Welcome to AI Video Mastery, where you will learn how to create AI videos that actually look real by combining AI with real footage. This class is for creators, marketers, or beginners who've tried AI before, but ended up with random clips that don't feel connected. If you want a clear system that helps you plan, generate and finish videos with confidence, you're in the right place. You'll work with JAPT and Clint to build a simple repeatable workflow based on three pillars, clear video planning, strong prompt creation, and blending real footage with AI. You'll create a finish blended video by generating scenes, editing them, or combining clips into a longer smooth final piece. I recommend watching videos in do because every lesson builds into the previous control the volume and the playback speed of every video tailoring at your own pace. If you can start you have any questions, be sure to drop them in the Q&A section below. Just make sure to check the existing questions first because there's a good chance that the question you want to ask has already been answered in detail. At some point, you'll be asked to leave review. Please wait until you've had a chance to really experience the material. Your honest feedback helps me improve the course and better serve you and future students. Thanks again for joining this class. I'm genely excited to help you create professional, realistic AA videos and to give you the confidence to use this system on real projects for the personal or clients based. Let's jump in to the first lesson. 2. Define Your AI Video Goal & Niche: Welcome. The first thing when it comes to combining real footage with AI is identify what exactly you want to happen. Do you want your character to change? Do you want the background to change? Do you want your character to do something unreal? Do you want the background to do something unreal? You need to identify what exactly you want, because whatever you identify will make it into a prompt that we will feed to our AI, and based on that prompt, it's going to do degeneration. But if we give it an empty promp, that's going to happen. So once you have a general idea of what you want, let's jump into the next video. 3. The Real + AI Video System: Animate from Last Frame or Image (Kling Workflow): Once you know what you want to generate, we need to go a little bit deeper and identify. Do you want the generation to start from, let's say, a specific frame after a specific frame of the video, or do you want to completely change the video and can change the video throughout? There's three main ways to go about this. We can turn picture into a video. We can start the generation after a specific frame of the video or we can change the video itself. And because of that, there's a slight difference in the approach that we use. For example, here's the frame of and you can just go ahead and animate this picture, animate this frame and do whatever you want to do. I can put this character into any situation then animate the situation, right? That's one way to go about this. Then we have this video of me running. And we can turn this video into something like, you know, the asteroids falling on Earth, or we can make sure that the asteroids start falling after a specific frame. And for that, we need to take one frame of the video. There are a couple of ways we can go about this. We can take a screenshot, but the screenshot has to be the exact size of the video. So if you have, you know, like, black edges on the side, it's not going to work properly. So you need to either crop your screenshot or you need to export a specific frame. In this case, I'm going to use Premiere Pro and was going to drag the video to Premiere and then drag it onto the timeline. And let's have me kind of in full size from this point. And I have this camera icon here. If you don't have it here, just need to click CPlus and then drag and drop it from here to here. But in this case, I'm just going to export the frame. I'm going to export it to the desktop. If you want to change the destination, just go ahead and browse and change it the filename and press okay. Now I'm going to have this frame that I will be able to animate with our AF. So if you have any questions, let me know that let jump into the next video. 4. How to Write Powerful AI Video Prompts with ChatGPT: Now that we know what's going to happen in our video, we have our frame or the whole video, we need to create a prompt. And you can generate a prompt yourself. However, if we have AI, why don't we use AI to generate a proper prompt? I found the best approach to work is to give instructions to something like HTPT or Cloud AI. Put your chaotic thoughts into it, it will put your thoughts together and generate a proper script. So for this video, let's just take a look at what I'll do with HTPT. It doesn't matter. JTPT clot or another AI. As long as it can generate prompt, it will work. And if you don't like the results, you can always just switch to another AI. Hey, JTPT, I need you to generate a prompt that's going to work with AI video editor or video creator. Basically, we're going to animate a frame of me running. I will attach the image for reference as well. Go ahead and take a look at it. What we need to do is we need our character to jump rapidly to the left into the water. Go ahead and create a prompt. Let's see what it does. See, it would definitely take a little longer to write this prompt than to say something like this out loud to GPTs. So that's why like using it. And if you read it, it's very well put together. Like, I don't write in this manner. I write in a much more chaotic manner. So this is really good. And now we need to generate another prompt. Okay, well done. Now, I'm also going to use a video, similar to this one. This is just a frame. And then for the second example, I'm going to use a video of me actually running. For that, I need you to generate a prompt where the person is going to run, but we basically have to change the surrounding area so that instead of water on the right, there's going to be a field of grass, and there's going to be like the forest instead of the water. So go ahead and give me the prompt. Very good. Very good. So once you generate the prompt, let's go ahead and jump into the next video. 5. Choosing the Best AI Tool for Your Video (Kling vs Other AI Options): You might be wondering which AI to use. There are so many different AI solutions. There's like pictures, images, there's a number of different ones into each category. Whatever command is a website called hicksvill.ai. It's a website that works based on credits. So you just buy specific credits or you buy a subscription that gets you credits, and then with those credits, you can use a lot of different AI solutions and test whichever works best for you. So if I go into video in the top left Korea and then go ahead and click on creative video, these are some of the past generations that I have. And if I click on the model, we will have an scroll down a little bit. We have all models, and so we have minimax, cling, and all the other ones. Here's the great thing about this website. It gives you, like, a quick description of what each model is good at so based on that, you can use each model. Different models use different amounts of credits. For example, my favorite one is Kling 2.6 because the results are really, really, really good. And for the videos that I did, it worked perfectly. So we have the motion with advanced video controls. There's just a new model that was released recently, for example, Kling 3.0, but this one uses a bit more character. So if I do 1080 and 10 seconds long. That would be 20 characters, whereas exactly the same model, but cling 2.6, that would take ten credits. And then if we use something like SRP, it's going to take even more characters, right? 12 seconds takes 150 characters. So to get started, I recommend cling 2.6, it's really, really good. And then once you get comfortable with cling, you can experiment with other different models. And overall, I recommend you Hicks fall because it just puts a lot of different AIs into one platform. So you don't have to buy ten different subscriptions. You can just get one subscription and test all the AIs that you. Then if you really want to, you can go ahead and then go to cling AI website, to the website itself of the cling and then buy credits directly on their website, and it's basically exactly the same thing. If you have any questions, let me know, let's jump into the next video. 6. Generate or Edit AI Video: Blend Real Footage with AI Scenes: Now it's finally time to generate our videos. Let's go ahead and do it. So I'm going to go into video, create a video, and here I will upload an image, which is going to be just the screenshot that we took. So I'm going to upload it here and I'm going to go to hATTPT, take this exact prompt, just copy it and paste it in. I'm going to use cling 2.6 10 seconds and click on Generate. For comparison, I'm also going to click and select cling for 5 seconds to see the results. And also the next thing is we go into edited video. And here we need to upload a specific video so let's go ahead and plot it. And then I'll go to Chat GPT, copy this prompt, and then literally just go ahead and paste it. And for this one, use a slightly different model, so you don't have option for all the models. And I'm going to go ahead and click on Create. And now, all we have to do is just to wait a little bit. As you can see, these ones are already in progress. All of them are now in progress. Sometimes it does take a little bit more time, sometimes a little bit less depending on how many people use the model, like how busy it is. Usually, it is always pretty quick. So let's just wait and see what kind of results we get. The first one is ready. Let's go ahead and take a look at it. Okay, the person jumping into water. Okay? So we definitely need to regenerate this one. Okay. Let's see what happens here. Okay, this one is better, so let's make it a bit bigger. So, this one is 10 seconds and this one is 5 seconds. As you can tell, the one that's a little bit longer is better, so go ahead and download it. By the way, you can also switch between the different viewpoints, so you can have a list or a grid. Let's go ahead and see this video. Believe this is the one. Interesting. We have the last generation. Let's go ahead and download it and watch it on the computer. Interesting. I mean, you can definitely tell that this is AI because if we zoom in, like, what's going on with my face? I don't have a face like this, and, for example, here, however, the fact that we can do this, a year ago, it was so bad compared to what we have here that in just a couple of months, it's going to be even crazier. So, if you have any questions, let me know, let's jump into the next video. 7. Create Longer AI Videos: Extend Scenes and Combine Multiple Clips: You might be wondering, how do you create longer videos by longer, I mean, longer than five or 10 seconds? Well, first of all, some models allow it. So if we go into create video and we switch to different models, we can see the length. So six to 10 seconds, here's clean 3.0 up to 15 seconds. So up to 12, this is up to eight, five, ten, and rock is up to 15. Usually, when we create videos, our videos are not super long because we do cuts. And so what works really well here is generating a number of different clips, putting them together in some premiere pro cup cut, whatever you use. So what I would do is I would go ahead and use the Kling 2.6. I generate a bunch of five second clips and then put them together and create a video out of that. So that's what I recommend you do as well. If you have any questions, let me know. But than that let's jump into the next video. 8. Last Step!: Congratulations to you. You are nearly 100% done with the AI video mastery. There are just two small steps you need to take. First, take action. As Kafuch said, a journey of 1,000 miles begins with a single step. So if you haven't already, take your first step by creating your first AI video. All the best information in the world means nothing if you don't act on it, and small steps lead to massive outcomes. Lastly, if you found involved in this program, I would really appreciate it if you could take 60 seconds to leave you honest review. I'll be immensely grateful to you, any feedback will massively help future students in deciding the best program for them. Although this course is complete, your journey has just begun. I'm excited to see videos online, so be sure to keep me and fellow students post it. Remember, I'm here for your success. So if there's anything you need, don't hesitate to reach out in the QN section below. In fatisic me as instructor, wishing all the best, and looking forward to seeing you in future courses.