AI for Architecture & Interiors: Image Workflow Guide

1. Ai Course Introduction: Tired of spending hours on renders and still not getting the visual quality your project deserves, there is a faster, smarter way, and it's powered by AI. Hi, I'm Amer Sawalha. And in this course, I'll show you how to turn sketches, cat drawings, mood boards, and ideas into stunning visuals in minutes, not hours. You learn how to convert cat drawings into photorealistic renderings, replace furniture with AI precision. Generate mood boards and conceptual sketches, transform two D inputs into three D models, edit materials, lighting, angles, and even remove elements instantly. Thousands of designers and architects are already transforming their workflow using these tools. Are you ready to join them? Join me now and future proof your design process? 2. 01_Flux kontext website sign up: Alright, guys. So the first software that we will be using, it's a website called Flex Context App, okay? And I want you to pay attention to the name of the app itself here. And if I want to sign up into this really, really great website, how do I do this? So first, I want to explore the pricing of what they've got, and they've got the basic P and MAX and they vary in the amount of renderings that you guys can do. So each rendering roughly takes five credits, and the basic gives you 100 renderings. This one gives you almost 250 renderings, and this one gives you 600 renderings, which is more than what you can imagine doing in one month. But I don't know. Maybe for large corporates, they really need this amount of renderings per month. So you've got it right here, and of course, if you need more, you can contact them and get more. Or you've got the one time. And the one times gives you 200 credit, 602,000 credits and each for really decent prices as well, but you don't get the access to more advanced models and features. So if you want to log in to this website or a sign up, it should be straight forward. You have to, you know, type your email and, you know, choose a password, and then you're good to go. Select a payment plan, pay for whichever you want to get as a package, and proceed forward. So for us to be able to apply on our examples, we need to render with flux. Of course, they give you four credits, but four credits is not enough to do anything because the minimum is five. So that's what I recommend. I recommend subscribing to this website, so we can proceed forward with our testing and our rendering edits. 3. 02_Angle Change Interior Front view: Alright, guys, so welcome to the first lecture. We will be discussing in this lecture the different types of AI websites that we will be using. So we will be covering almost as mentioned in the introduction, 20 architectural interior workflows. And these are very beneficial for us to be able to move really forward with our concepts, with our renderings, with our CAD, so many things. So first thing first, I want to talk about the flux context. So the Flux Context guides is a main website, and it can be integrated into other websites. So I can go to other websites and find flux as an option for us to be choosing among. The one I'm using now it's called fluxcontxdt app, and this is the one that I'm currently using. Flux Context is produced by Black Forest Company. And as I mentioned, if you go to Google and type in Flux context with a K, we can find so many websites that is delivering the flux context. This assures the idea that I mentioned to you guys, is that it's available in so many websites. The one that we will be using, though, is called the Flux app. So make sure that this one it is, so you guys can follow along and make sure that you are using the same website with the same tools and everything. It varies from a website to another. But for now, we will be using this one, so make sure you use this one. For flux contexts, it needs a subscription, guys, for us to be able to move forward with it. It's a cheap subscription. We can use it for so many tokens, more than what we need, and they've got different types of subscriptions that we can enroll in. It depends on the activity that you guys are planning to do. All right, what does Flux do? So basically, here's some explanation is that B image contexts from A. The bird is now sitting in a bar and enjoying a beer. So this is the rendering that is generated, and A, assuming that it's the bird only. Then you've got C image context from B. So this image from this one, there are now two of these birds. So the prompt is there are now two of these birds, and these birds are right here drinking beer. Again, we've got D. So from C, watch them from behind. So now, since we generated two birds, we got two birds from behind. And E from C, the two birds characters are now sitting in a movie theater. So I think you guys now grasp what does flux Context do? It edits images, but with preserving the original image. So this is like the revolutionary part of the AI is happening is now usually AI guys renders something with the same idea, but it doesn't render the same concept and the same output. So in this case, we've got a website that basically renders someone calls it 90, 95%. But in my perspective, this is very good and can be used in our workflows. So what can we learn from flux? So there are so many products we can do from flux. Some of them is doing interior angles, exterior angles, image edit, inserting materials, element addition and removal, day and night and close up. Also, we will be discussing in a GPT, what we will be doing. So in GPT, we're going to do CAT to Maquet, CAT to rendering, CAT to sective. We'll be doing elevation to CAD, empty room plus moodboard, furniture replacement, furniture to CAD, moodboard extraction, and render to sketch. And I've got some image editing, we will be using rea. So we'll do Crea for concept generation and image enhancement, and I've got some others, I'd actually call it bonus for us to use image to three D, so we can now generate a three D model from an image and watermark removal. So you guys can imagine how much we can do with these tools. So first thing first, we'll do the flux, and I'm going to start with the interior angles. So I've got this image right here, it's surrendering. And I can see that we've got the bed from an isometric view, and we've got the windows. We've got also some walls, and we've got some tiles here as well. So I want to be able to create a different angle from this image. Revolutionary. Right. So let's jump into it. So what I will do is that I will drag and drop my image here, and I've saved the prompts for you guys, so you can copy and paste the prompt itself. So here, Control C, Control V, and I'll do generate a new view of this render from a front elevation where the camera shows the bed from the front, and I can add preserve all details. And I'll do create. And as you guys can see, each image requires five tokens, and if I enable add more images, such as here, so I need ten tokings now, but we're going to add only one image. And what's great is that the aspect ratio, I can match my aspect ratio. So I've got the aspect ratio, I believe, 16 to nine, and it's already going to match. But I have the flexibility to be able to use whichever aspect ratio I want to use, and there are many. So whichever product I'm rendering for, this is the great website to be able to use because I don't want to just be able to create a front or higher elevation view. I want also to be able to control the aspect ratio because it's very important. So I'll do Create now. And I'm going to wait for it for it to render and see what I can get. And here, as you guys see, it's already using the tokens. So that's a front elevation of the same rendering, and it's helping me a lot visualize my design. The way I want it. So now I can download this and guess what it downloads with pretty decent quality. So if I double click on it and I zoom in, I can see that the quality is very good. Yes, if I zoom in a lot, I've got some pixels here, but guess what? We can enhance this. So this is not a problem. But we all know that from any three D software, whenever I'm rendering, so I need to at least if I'm using a real time software, which is revolutionary as well, I can just insert the camera and render it, but I have to wait. In all softwares, I have to wait. But in this AI website, I don't have to wait, and it preserves pretty much most of the details that I have. So this is great. I'm going to do Control X, and I'll do Control V, and I'll do here zero, two, underscore, and I'll go here, copy my prompt, which I added, by the way, preserve all details with it and put it here. I've already got a rendered image here, but the prompt does not say preserve all images. So the AI suggested that I have a window here and it's showing the rest of the view, but this does not go the original design, right? So I need to do it in a way that is as close as possible to the details that are available in my design. So this one does not have the preserve all details. It has generated, let's say, somewhat called 90% close image to what I'm looking for. But with the prompt adding that preserve all details, now I've got all the details that are already in my original design. So you can see that this is the power of flux context. So I will be distributing the lectures among each exercise so that we have short lectures and concise. I hope you guys really benefit from this one and see you guys in the next ones. I'm really excited because what's going to happen next is so many life easoning tools. See you guys. 4. 03_Interior front view result enhancment: Okay, so in our previous lecture, we did a front view for the bed from sort of an isometric view for the bedroom. So I've got a chapter that I will be talking about image enhancement, but I want to jump into it a little earlier because I want to give you a perspective on what you guys can do in your workflow. So for instance, here, I've got the rendering that I have, okay? And I want to somehow make flux give me a better result already, right? Because we all agree that when we zoom to a certain limit, it starts to pixelate, even though it's like a 1.5 something megabyte image, even here, so I can do either or I can enhance the original image and generate the new one with an enhanced result. Or I can take the enhanced result. I can take the one that is generated and enhance its result. So let's see together. I've got a website here that is called crea.ai, and it's pretty famous in the AI websites out there. So we need to do an account in Kea, and Kreas considered somewhat a free website because it has a subscription, but it gives you a lot of opportunities to be able to test with so many features before you decide to subscribe to this software. So I need to sign in from here, I'll do Logout for you guys to see how you can create an account. So you can do a signup when you do a sign up, it's going to ask you for your email. It's going to ask you for a password, and it will ask you for verification at best and you can come back with the code and then you are already created an account. But we'll do a login for now and I'm just going to log in with my email and it's already has saved the password, and maybe it's going to ask me for some verification here. I'm already logged into Crea. So this is the main interface of Create has so many products. The one that we will be using is enhancer. So we'll open the enhancer from here. And, you know, I've got the history that I've done already some tools on the image. If you want to do video enhancement, if you want to do real time, edit, and if you want to check assets. And here tells you upscale and generate new details for images up to 22 k. That's a huge number. And videos up to eight K. That's also a huge number. And you guys need to consider that you need to subscribe to be able to get to these numbers. I'll do Upload, and from flux here, I'll go to interior angle, and I'll choose the latest result that I was able to generate from Flux. And I'll do open. It's going to take some time. All right. And then here it gives me more things that I can control. So AI strength, resemblance, clarity, and sharpness. And the more I play with these numbers, the more I'm going to get different results. And here I can type in some prompts to describe the image, so maybe I can get more accuracy. But what I will do for now is just enhance because the defaults are already great, and I will see what is the result that's going to come up with. I'm really excited about this because crea.ai is one of the really good websites that are free to use, and it can generate really, really good quality renderings, especially if I want to enhance them. So let's see what can happen now. Great. So it has already enhanced the image and it gives me somehow before and after, and you guys can already see the result. There you go. So now the result is much better here. This one looks pixelated compared to the new one, even the greenery. Look at the greenery here. It's really awesome, guys. I even love the rug, the details. It's really cool, really, really cool. So now I can consider that this is the final output image, for example, for the one that is generated from Flux. So I can take the edited one from Flux, put it in Crea, and now I can download my rendering and I can do I can take this image, Control X, and I want to go to where I have my course, and I'll go to flux interior angles, and I'll do Control B here. But what I will do is I'll do 03, and then I'll do here, underscore enhanced. So you see how my image is evolving the more I'm adding prompts. So here I ask for preserved the details. Therefore, I got a preserved detail shot, and now I've enhanced this shot to make it really stand out as an ultra realistic rendering. So all of this combined guys can give us the work that can save for tens of hours, literally tens of hours with no joke. I hope you guys benefit from this lecture, see you guys in the next. 5. 04_Change Angle Exterior Aerial: Alright, so now jumping to the next exercise is that we've got this exterior image really wonderful, and we want to be able to make two other views. One is from a bird's eye perspective, and the second one is a front perspective. This is the bird's eye, you guys can see that if you want to achieve this on three DS Max or any three D rendering software, it's going to take us so long to be able to do it this way. But in AI, it's doing it in seconds. Yes, I agree that there are some imperfections, but according to the time, this is wonderful and any other unperfection can be fixed. So this is the front view. So we'll see together how to create all of these. So we'll go to flux, and now we're going to remove the previous image. We're going to do this image. I'm sorry, we're going to add this into the drag drop. We're going to delete the prompt, and then we're going to take this prompt is change the view angle to an aerial view control V, and I want to ask it to reserve or image details. Let's see, it's still uploading. I think that image up to 5 megabytes maybe image here no image is fine, so it should upload. Guys, anything in this course that will take long time with you is that this could be because of your Internet connectivity. So the better Internet connectivity that I have, the faster the process will go. So I'm going to just, you know, reload. I'm signed out, so I'll sign in now. It's going to assign me automatically into my account, and I'll import this one, and I'll take this prompt, Control D. Either the Internet connectivity is low or I need to refresh the page. Aerial view and preserve all image details. Generate. Let's see what flux will give us. Great. So the image is rendered. I've got all my details preserved. I'm going to download this one. And I'll do Control X. I'll come here, ControlV and I'm going to compare between this one, which has preserved most of the details, as you guys can see. And here, this one hasn't preserved all the details. So the word preserve all details is very important because it tends to help us get the same details of the original image. So technically, I've got three houses here, and this image, because I haven't prompted preserve details has one, two, three, four, five but this one here has one, two, three, so it has preserved all the details and already had rendered this house for me here. So preserve our details is very important for us to be using whenever we are prompting. Okay, good. So now I will take this prompt. And I will put it here. So I will be prompting from the original image, right? So if I want to edit on anything in this image, then I upload this image and prompt it again, which we will see in the materials edit section. But for now, change the view angle to a front view and preserve all details. Generate. So the difference between this one and this one is the preserve all details. So Control V and preserve all details. Great. It has gave us something, but not the thing that I was hoping to get. It should be something similar to this. So what I will try instead. Of course, if I keep on trying, it's going to keep on rendering different images, front view and preserve all details and building orientation and spacing. Let's see what happens if we generate again now. Is it going to be accurate or is it going to require me to give it the top view, the aerial view, and it's going to generate the front view from the top view. So now we'll see. So it has generated something, but it's not the same that I want. Sometimes the website, I need to, you know, refresh the website, so I'm going to Control C, refresh the website, upload this image, and I'll do this prompt. Sorry. So it's going to be change the view angle to front view. Pretty simple, pretty straightforward. Let's see what it can get us generate. So, guys, there are some flux websites out there that is part of other websites that can generate up to four images in the same time. So you guys can check and see rather than waiting one by one, you can wait up to four together. Okay, so it has gave us something. It's not the same thing. So maybe it needs more data details. So I'll upload this image instead. Wait, let me do the exact one. Generate the view angle to our front view. And let's see what's going to give us. So it has uploaded, and I'll do Create now. Wait to see what we will get from flux. I already has rendered this one. But let's see. This is somewhat convincing one. Maybe we can change the view angle to front view elevation or orthographic elevation. I'll take this one and I'll do create again. So till it's render, I'm going to go and check it out. Pretty good. I can prompt for having a human eye shot. So not to take it from Avove, but this one is also pretty great, pretty straightforward. Again, this is AI. We can't expect from AI to understand everything. Sometimes we need to really be detailed with it for us to generate really cool renderings. So this one is good. I'll remove the front view and I'll do human change the view angle to front view human eye. Let's create. So I'm doing this on purpose with you guys to see what can we stumble upon while we are in this process and how can we work around it rather than just giving you the image itself and how it turned out without understanding really the logic behind it. So it has gave us something that is front but not really front. So the human eye didn't really work. So, again, I can refresh the page and upload it again. And I can prompt the same prompt and see what other result can flux give us. So control, we generate. Yeah, I'm spending my coins on you guys. Yes, for us to be able to see really, how can we get the desired result as close as possible to what we're looking for. So now it has gave us something. I can add preserve all details to see what we can get. I like this one, as well, but I want to kind of give it like the really front view of the three cottages here. Preserve all details, generate. All right, so it gets us somewhere close to what we are looking for. I'm pretty sure that if we prompt even more, then we will be getting the exact result. But you need to know that, again, this is AI, so it will get us as close as possible to what we are looking for. Course, maintaining simplicity with a prompt and being straightforward is very important so that we can kind of achieve we can kind of teach the AI that we're looking for exactly this result or that result. So I need to keep on testing and prompting till I get what I'm really looking for. And there are some different applications that I can do through. So this is to give you a glimpse of how wonderful results can be created through just one image, and I can create, like many renderings for the same project, and it's talking the same identity, talking the same visual fidelity, and stalking even the same design that is out there. So I hope you guys benefit from this lecture, see you guys in the next ones. 6. 05_Close up shot: Alright, next. So now we will talk about close up shots and how to create close up shots. So I've got this rendering, and I was able to achieve this close up look from Wonderful, guys. And I've got this image, and I was able to generate this close up shot. So for this demonstration, I'll take this image, and I'll show you guys how to produce this close up shot. So same thing. We're going to go to Flux Context, but this now, I will reload and refresh the page. So just to make sure that it has cleared the cache of previous workflows. So now I'm going to drag and drop my image, and I'll take this prompt and paste it here, which states that make a close up shot for the vases with a depth of field blur to everything else except the vases. Generate. And there we go. It has generated the image that we are looking for, and it's pretty great. So I'm going to download this one. And, of course, guys, any image that I can take from Flux, I can put it in Crea and I can generate a higher quality one. So we will talk more about rea as we've talked before, but we're going to do a couple of other examples, not this one, but for you guys to grasp on how can we do a close up shot. So I'll do another example. I'm going to remove this one. I'll put this one, and I'm going to copy and paste the prompt because I made sure that I'm describing the image in each prompt. So for now, it's the juice jar, not the vases with focusing on the jar and blurring everything else. We can even focus on something different now. So for example, I want to focus on the shelves, okay, and plates in the shelves. So I'm not sure how is it going to look because the background is going to be really busy with the plants. I can Describe, let's see, I'm going to do the plates for the demonstration of that. It focuses on what I ask it to focus on. So make a close up shot of the plates on the shelves. Focusing on the on the plates and blurring everything else without the coma. Let's see. Okay, I did. I did. Perfect. That's great. So I can take this image. I can put it part of my mood board. And whoever sees this image will be able to tell that this is here. Around here. So this is great. This is how we can make a close up shot. Okay, I'm sorry, I mentioned it close up shot. Okay. And I can have details such as blur and high quality focusing. So this is revolutionary, guys. Usually, I need to create a camera inside TDS Max, put all the depthal field and, you know, the aspect ratios and the aperture and everything for me to be able to get something that is this good. So now we've got a tool to make close up shots. I hope you guys benefit from this one. See you guys in the next ones. 7. 06_Night shot from a day shot: Alright, so talking about something new, a new workflow, a new application, we will be talking about day and night. So this is the day shot, and this is what I was able to achieve in a night shot and even this one. So each one differs. I'll show you guys how did I prompt this one. So I'll remove I'll do refresh page. This is what I recommend personally, so it does not take anything from the previous workflow. I'll upload this image, and I will take this prompt, make this a night shot, and I'm going to show you how we can tweak our night shots to get as close as possible to what we are trying to achieve. So I'll do generate now, and I'll see simple make this a nightshot. So sometimes I need a complex prompt. Sometimes I need a really simple prompt to describe what I'm looking for. Because the more I complicate it, the more I'm going to, you know, get this jittery results that I'm not looking for or this additional objects. So it's not really always necessary to be specific, detailed. Sometimes it is, but sometimes, you know, having a simple prompt can do the job. So there you go, guys. We've got a night shot, and if I want to post produce this one on Photoshop and be able to achieve just this one without editing it, without anything else, then it's going to take me a good amount of time to be able to, you know, have the lights, you know, extract the sky is the easiest. But, you know, have the GI effect on everything else. This is going to take me a decent amount of time to be able to produce a night shot. But in our case, it was a couple of seconds. So, great. I like it, but the problem is that I believe it's a bit dark. So what I can do is that I can take a new prompt now with the difference the differences here is that the shot with a high light intensity, also add lighting to the street polls and a blue sky. So I'll do Control D here. And what I will add different since we're experimenting together is that I want to go and tell it that I want also Light the neighboring building. Let's see what we can get. So when I prompted that I want even higher intensity and light up the street, this is what I got. This was the previous one. I got a higher intensity one and even more beautiful one and more realistic. But now I want to tell it to light even this building. So let's see what's going to come up with. Great. Great. It hasn't lighted the building that is near to it, but it has definitely light up the street, and it has came up with really beautiful colors for a night shot. Really realistic. So this is the last one I was doing on my own, and now we've got this. So to compare, they're very close, 'cause, like, it's the same prompt, same image. But I believe that this is less saturated. I like this one more. So you guys can see with just a couple of clicks, now we can get a night shot. And usually, if I want to do this in a three D software, I need to flip the entire setup and the entire lighting and assign lighting to objects and do the ambient light for us to be able to get night rendering. And usually when I'm doing this, I need for me personally, I take a copy of the file and just name it night shot to be able to, you know, produce as many nightshots as possible, rather than, you know, flipping between day and night all the time to render day then night, then night then day, and it takes a lot of time. But this one takes really, really short amount of time. So this is how we create a nightshot from a Dayshot. I hope you guys benefit from this one, see you guys in the next ones. 8. 07_Add and Remove elements to our rendering: How can we remove all the furniture that we have in the interior and get a clean interior space? So for the future, in another tutorial, we will learn how to get a mood board and furnish this space. Or this is called element removal. We're going to be also discussing Pardon me a second. We're going to be also discussing how to make element addition. So how can we add an element to our rendering rather than deleting it? All right, so I will refresh flux, and I will take this rendering, put it here. And I'm going to prompt empty all room furniture like this. I'm just going to remove the file numbering. Empty the room of all furniture. Create. Okay, so this is what I got. You know, it has removed only the chair, but I will remove empty the room from all furniture and see what we will get. And this is what we've got. You see, guys, just a small word affected what we get. So don't get frustrated. You know, keep on trying and save your prompts. So every time you want to just do what you have in mind, you've got the prompt for it, so you don't have to waste credits for it, but it's fine. For me, I am using my credits for you guys to understand how can we deal with AI. So that's the first one on how to empty the room from all furniture. Let's see how we can add to our room. So now I will remove this one. I'm not going to download this because I've already have it here, and I'm going to delete the prompt. And from here, I'm going to put this image, and I'm going to ask it to add an Abs carpet to our interior. I'll take this, copy, paste, add an Abis carpet horizontally on the floor between the two hallways matching the vibe of the interior without the comma. So generate. And there we go. Now we've got a carpet in the middle of the hallway. The difference between this image and this image is that I just prompted for the carpet differently. Here, I asked it to have a carpet horizontal on the floor between the two hallways matching the vibe of the interior? No, but this one I ask for an abstract, simple white carpet. So I'm going to do it for you guys. An Arabisq. It's already here. Add a modern white Arabisqu. You guys see. So this is the only difference that I've done to be able to achieve different carpet. Look. And you can measure anything upon this. So you can add a couch. You can add a chair, you can add walking people. Whatever you guys are aiming to achieve, you can achieve it through prompting it correctly. 9. 08_Edit our rendering material and adding people: How to change this material in this image to this material right here. So I've changed it from bricks to concrete with just some prompting. So let's explore together how powerful this is. So I'll refresh flux context and sorry. And I will navigate to where I've got the image, and I'll drag and drop my image here, and I will ask for change the bricks into gray precast concrete. And generate. Let's see what we can get. Perfect. So we've changed the entire material of the whole image with just one prompt and maybe 5 seconds. Usually, if I want to do this in a three D software, I need to assign this material to every single object, make sure that the UVs are correct, and then rendering the image, and maybe I can get some mistakes and re render and so on and so forth. Then I need to put it in a retouching software such as Photoshop and edit it. Okay, great. So we have changed the whole scene material. What if we want to add more people? So if I want to achieve this, I need to put it on Photoshop and add person by person and then do the motion blur for the person, and then after that, do the shadowing of each person, which can take really long time. Maybe it can take up to an hour to get it with this quality and the seamlessness. With our tool now, AI tools, especially flux, we can achieve this. So what I will do is I will remove this image because I want to update this one now, I'll put this one, and I will prompt for Make it busy with people walking, add Motion blur to them so they don't look unrealistic because motion blur I AI is your best friend whenever you want to add people. It can add really sharp people, but then you have to enhance them a couple of times to get them with the correct faces. But when you do the motion blur, it kind of helps the AI to generate a realistic one. So I will create now and see what flux will give me adding people and adding motion blur to them. And for the result, there you guys go. I love it. I literally love it, guys. We'll download it and see how's the image. There we go. It has added people with their shadows. They look realistic, and they look as they are, you know, has been shot with a camera. So I just saved in this video alone around two to 3 hours of work in just 3 minutes, literally. So using AI is like 1% of the time that I need to do to be able to achieve the same result. And maybe sometimes AI can give us even better results than the one that we get. Of course, if we are professional three D renderers and three D modelers, we can achieve amazing results. But compared to the timing, this is like a whole cutting edge different story. I hope you guys benefit from this one, see you guys in the next. 10. 09_Assigning Materials: Now we will talk on how to change our white flat three D model into a fully materialized rendering. So you guys can see I've already have materials, I have lights, I have a really nice shot. Even I've got fair chairs, and usually if I want to achieve this, it's going to take me a lot of rendering time, and I have to have an exceptional CPU and GPU to be able to achieve this. Taking you guys to the timeline, this was this, this, this, and then latest this is the first thing that I began with. Then I've added some materials, then I've added the carpet, then I've added even more materials, and then I've added lighting. So how can we achieve this? So I will take the image itself and I will put it here, and then I'll ask it for this prompt, which states that make the floor epoxy concrete, counter white marble, table wood, cupboards, white laminate, glass lights, and burgundy for the dining chairs. Great. Burgundy fare for the dining chairs. So I can do it this step by step, or I can have the whole prompt going on. So what I've added here is that I've talked about the carpet. So here is also abstract, and that is white. Here, make the circular rug under the dining table one. So I can add this space. I can add this as well. And what did I do different here is that circular chandelier, make the circular chandelier glass sphere with lighting inside them, turn on the lights of the spots, and the group fall ceiling, make the accessories on the bright, silver, blah, blah, blah, blah. So I can even copy this one and I can Enter and I can paste. I can put all my prompts at once, or I can do one by one. So we'll test first is that to do all in one, again, I'm going to mention to turn on the lights and create. Let's see if it can achieve all my thoughts with one trial. Then I'm going to save you guys the time of doing three or four trials to achieve whatever we're looking for. V, I did with a big percentage, it did. This is what I have now. It's really great. It's really wonderful. I can add this and continue prompting. Let me delete this one. And ask you to turn on the light in the hidden ball ceiling groove. What else? Make the fridge on the left silver metallic and make the cupboards reflective laminate. What else? Give variation to the books and accessories. Give a material variation for the books and accessories. Let's do create and see what we will achieve. There you go. So it has let me download it. It has lit the false ceiling and has added the metallic fridge. It has made the cupboards reflective and has, you know, brightened the image a little. So to put them next to each other in context, this is what I have now. So it has changed the fridge, and it had turned on the fall ceiling light, and it has gave maybe not much focusing on the books. Maybe if I take the image itself and just prompt for the books, then I'll be able to achieve a variation of the books, but I love it. I love it. This was this, guys. Literally, this is revolutionary. Usually I need to do it on Photoshop. And if I do it on Photoshop, then it's going to look quite close to this one. And even if I want to do it on three D and be able to show the progress, for example, for instance, I want to do an animated video on how did I materialize every single element on this? So I can prompt for element by element and then after that, in the end put them in a sequence together and show how did I change the whole material of everything here. So this is really powerful. This is one of the applications and the workflows. I hope you guys really benefit from this one, see you guys in the next ones. 11. 10_Rendering or photo to a real 3D maquet: Alright, so we're going to talk now on how we're going to do this aerial image to three D maquete that we can sell for real estate properties and developments. So how can we achieve this? We're going to refresh flux, and then I'm going to put the base image, and I will take this prompt, copy it and paste it here. Make this a three D Maqette model for a real estate showroom, make it look like a real generate. Let's see what flux can give us. And there we go. That's a three D maquete. So just with a click of a button, we can achieve a really good result. And with few more prompting, we can remove whatever is extra and that we don't need or want. And of course, we can enhance this image after. I want to show you guys how the enhancement of the image can be done, so you don't have to do it. So this is the base image or this is the generated image, and this is the enhanced one. Not much of details. I believe even the enhanced one has hidden some details. And, of course, the three D Maquetes usually take efforts for it to be perfect, but I can tell my idea, right? If I'm, you know, debating this with a customer, I can just show him and be like, do you mean this, and they'll be able to tell me yes or no. When usually I take whole longer time to be able to actually generate this and convince them of the idea. Now we're going to apply the same thing for a house, a CAT plan. So this is a CAT plan, as you guys can see, and then we're going to turn it to this rendered three D form maquet that it's like a cross section for the house, that we can see it from a plan view. So how do we achieve this? We take this prompt Control C and then Control B here. Make this a three D Maquette model with extruded walls and windows placed in a meeting room with a person pointing at it. Let's see what flux will give us. There you go. Amazing. I can see that our ambitious cat plan has been turned into a three D Maqute that I can as well sell it to clients or real estate development companies. I don't know how long did I save you guys in this video, Usually days, but I hope you benefit from this one and see you guys in the next one. 12. 11_CAD to Rendering: Hello, guys. So we finished the first chapter, where we talked about flux and now, you know, diving even deeper into more architectural and interior design workflows that would save us a lot of time. So we're going to talk about Chachi Pit now, and most of us know Chachi PT for writing emails, fixing contexts, applying for a job. But now with the latest updates, four point oh, we've got, you know, changes on a whole different level. It can save us thousands of hours. So let's jump into the first application of this, and we're going to do CAD to rendering. So basically what we will learn is how to have two plans, convert them into a three D realistic rendering, such as this one. Who said that one day we will be able to achieve this from Chachi PT and even have an aerial rendering of it and can enhance it later on. So what we will do is that in CHATT, what I recommend you guys is to buy the plus because if you don't buy the plus, then you'll be able to generate one image per day, which will take you months to be able to finish a project. So rather, when we have the plus, we have the priority, and it's much quicker and it's infinite amount of edits and renderings. So that's number one. Number two, Focus on this point, guys. Please, whenever you want to achieve, for example, things that are relevant to the same project, then we do it in a chat. If you want to achieve a new result or a new project, we do it in a different chat because then ChachiPT will remember the old chat and start combining two chats together. So each project open it on a new chat on hatchiPT so that you guys don't mix results with older projects. So I'll take these two images drag drop here, and I'll copy and paste this prompt, which states make an editorial photographic ultra realistic three D rendering for this plan and elevation. Note that this is in Hajia Sofia Mosque in Istanbul, Turkey. So I've also give Chat ChiPT the landmark that I'm trying to render and where is it located? So it will now combine real world data with plans that I have and try to mix them together. So if I know the project, it's even better. And now I do enter. So I wait for haipit and see what Cha hipit will give me, and now it's generating the images. Usually, Chachipit takes more time than Krea, but if I have fast Internet and I've got plus, then there should be no worries, you know, achieving a much quicker result. So it's going to take a bit of time. So I'm pausing the video for you guys to not wait so long rather than jump to the final result directly. And Bola. We've got our image, and now we can download it and check how great it is. I mean, come on, guys. This is amazing. This is, like, a high fidelity rendering that if we need to do this on three D model and render it's going to take us a lot of time and even look at the variation of materials and to achieve them to make it look this realistic. This is brilliant and amazing. Okay, great. So now we will ask it to give us an aerial view of this. So I'll jump back here and give me a bird's eye. Enhance, give me a bird's eye. So, okay, give me a bird's eye. No need to tell it for this because it's already open in one singular chat, and we've done one prompt for the same building. So no need to, you know, even highlight more things unless I want to, for example, tell it to show me the garden or the existing context of this building or, you know, make it look like a building in the middle of the desert, stuff like this. Then I can prompt even more. But if I want just the perspective of this, then I can just give him give me a bird's eye or give me an aerial view of this rendering. And there you guys go. We've got Bird's Eye just with a click of a button for the same structure, for the same everything. So this is really powerful whenever I've got. You know, I finished my planning. I've finished somehow architectural drawings, but I want a three D rendering. So, you know, this is amazing for the quick result that it gets us, guys. Usually, the quickest result was back then was maybe Lumian, but the result looked really horrible at first because you need to, you know, if you want to get a really good result from Lumian inscape, these real time rendering softwares, then you have to put the same amount of work, the one that you do on three DS Max and any other, you know, unreal complex softwares, but really high fidelity software. So now it's a game changer. Now, if the client is coming to our company and we need to show him something ASAP, then somebody can go inside a laptop and just generate renderings. This is revolutionary, guys. I hope you guys benefit from this one, and I'm really excited to share with you the rest of what we can do. 13. 12_CAD to Sective: Alright, so now we've got a CAD image, and we want to create a sectv with worshippers inside, assuming that this is a mosque, or I can do a sectv in any building and just give the activity of the people. Having that this looks similar to what I was resulting here. So what I will do is that I will drag and drop my image here, and I'm going to copy this prompt. And basically, the prompt says, Make an editorial photographic ulterial three D rendering. Cut sectiveOh, I'm sorry. Cut sective showing worshippers inside. Let's see what Cha ChiPit can give us now. Alright, so this is what we've got. And this is also really cool and really interesting, and I love it, but this is not what I'm looking for. One thing, guys, you need to consider when dealing with ChachiPiT is in flux, for example, it had the option for us to match or change the input of the rendering. But in ChachiPT, it suggested itself, that this rendering should look like this, which is not what we're looking for. First, the result is not what we're looking for. And second, you know, it hasn't preserved. You see every time it's giving us a different ratio. And maybe this is good, maybe this is not for us. So just in case we need to be specific about the ratio that we want Chat GPT to render the image with. So now, it has gave us interior of the highest Sofia mosque in Turkey, you know, because we started the whole conversation about this, you know? So if I want to have the sective in isolation of what I've already done, then I suggest that we do a new chat, and from this new chat, I can do this CAD and I can Control C and control V my prompt. And just generate the prompt and see what we get out of this. Again, I love the interior option that it has gave me, and it saved me a lot of time to render this. And usually, I need to look for people that are worshipping either three D model or I can look them online, and then I need to arrange them. So it saved me tons of time already. But in our case now, we want just the sective of the rendering. So I've done a new conversation to do it, and I'll wait for it to finish. After this, if I don't like the aspect ratio, I can also change this. Perfect. This is what I've got now and I'm going to download it, and it is the same as the sective. So as I told you guys, whenever we have it as a different conversation, then it will give us the result that we're looking for. So let me refresh, and this is what I have, and I can even enhance it further. But easily, if I'm an architecture student, I can put this as a sective and just have the lines darkened and there you go. It's a rendered sective. Usually, I need to do this as a three D model and then cut it to be able to achieve a sective and it's not even rendered, then I need to render. So you guys can imagine how long it would take. So I hope you guys benefit from this tutorial. See you guys in the next ones. 14. 13_Elevation to CAD: Guys, interior designers go to love this. So how can we achieve an interior elevation cad elevation from a three D rendering image? So, pretty simple, pretty straightforward. This one is tricky a little, so we need to focus and follow along. Great. So I'll open a new conversation with ChachiPT. I'll upload this one, and I will do Convert convert the interior perspective view into two DCAD elevation, without any perspective lines using clean black outlines, keep all design details without any changes. Let's go. Let's see what we can get. So I hope that it changes the result or it reads the result well, but the first result that I got from Chachi Pitti was this one. Even if you get this one, don't give up on this guys because you can even, you know, prompt different to get this result. And I will show you what you need to tell ChachiPit. But the thing is that, you know, you got to take it step by step. So as I told you guys, it has gave us the exact image that we are not looking for. So what I've done is that, you know, I've talked to Chat ChiPT even more, and this is what I told it. You know, I can easily give you the prompt from the beginning, but I want you guys to understand that this is not perfect. You need to find your way around to be able to really make it understand what it needs. So I told it, No, this is very bad. I need a much higher quality one. Besides, this is a three D trace. I want it to be a two D and no need to draw the tiling, type in tiles and add dimensions. Let's go. Let's see what it will come up with. So now it's going to analyze it even more, you know, and see what it will give us. So this is what we got now. It's a much cleaner and higher resolution, and it talks more logic than the one that we had earlier. And of course, we can submit this as a CAT, but we can't submit the previous result as a cat. I can't go to a client and be like, There you go. That's a cat that you're looking for. No, this is not. And, you know, as I mentioned, it's called a three D trace. What we need is a two D cat. Okay, great. So we've got this, and I'm going to download it. So now we're going to jump into the point where, how can we convert this to an actual cat? So I will go into a new tab and ask for Convert To image to CAD AI. So there is a website, free website that can convert our JPGs to CAD. So JPG I'll pick from here CAD and I can do DXF, and I'll choose File. I'll go to the downloads and I'll make this bigger. I'll take this one, and I'll do Convert. And let's see Convert result. So it's converting and it's going to take a little while for it to convert. Not so long. Again, depending on the Internet speed and depending on the complexity of our drawing. Now we can download this file, and we've got our downloads here, and we can double click on this one and open it with Autocad just once and see Autocad will read from this converted file. Okay. Z O. So this is it. So I press Z then A Z for Zoom, and then A for A. When you click Zoom, it's going to ask you all center dynamic extents previous scale, et cetera. So I'll do A, and then I've got our drawing and it is CAD. So this can help us a lot, even when we're doing three D modeling. Is that I'm generating a cat or the base cat from the image rather than, you know, drawing on top of it from CAT and then taking it to, for example, three DS Max and extruding it. Usually, I can also generate details from this process. So this process saves us a lot of time, guys. I'm going to call this course the time saver in AI. I hope you guys benefit from this one, see you guys in the next. 15. 14_Empty room + mood board: Alright, guys, so now we're going to talk about how to do this mood board into an empty space through HathiPT and actually generate a three D rendering out of just the mood board and an empty space. So the empty space image I've got is this one, and now I'm going to open a new conversation, and I'm going to give ChachiPT these two, and I will copy this prompt where it states the following. Furnish this empty room with mood board items, preserve the camera angle, proportions and existing openings materials, openings and materials. So now Chachi PT will combine this space and the proportions of it, the aspect ratio, and we'll add the elements of this moodboard into this rendering, resulting in this image. So let's wait for it till it generates our image. As you guys can see, it has already generated the image. I can download the image now and check how is it looking. And there you guys go. It has rendered our mood board inside the empty space with proportions and aspects of the same this area and give us a really clean and neat rendering. And if we need any other details, we can just specify and ChaiPT will generate it for us. So there you go, guys. Now you can do a three D rendering out of a mood board. No need to put them inside DS Max and render them. Of course, that's a different pipeline, but now we can generate it in seconds. So this is really powerful. 16. 15_Furniture replacement: How can we have these two chairs replaced by these two chairs and maintaining this rendering. So this one is furniture replacement. So I can love the shot, but change a couple of things inside it. No need to change the whole thing. How can I do that? So first thing first, guys, I've managed to mark the image here, and how do I do this? Is that I double click and I go to edit, and then I go to Markup, and I can mark the objects that I want to replace. Then I can do save options and save a copy, and it's going to ask me where to save, but I got it already here. So I don't want to save it. I've got it already marked for me here. And I've also marked the chairs that I'm interested in changing too, right? So now I'll take these two images, I'll open a new conversation. I'll take these two images and I put them here, and I will copy and paste this prompt, which states, replace the two orange chairs under the stairs with the two gray chairs and replace the exterior view with the city of Hamburg Skyline. Let's see what we can get. Lilly Renders, I want to show you guys how it turned out. It turned out exactly how I wanted. So it replaced the gray chairs with the orange chairs, and it has given me somewhat a skyline of Hamburg or any other view that I want. And Voila. There you guys go. We've got our rendering, and it's exactly how I want it to be. But the thing is that it's imperfect. Why? Because I've got, you know, the skyline is already here. So I can tell ChachiPT that no, or I can tell it. It's great. But the skyline is over laying the interior of the building. So concise Concise it to the window only. And of course, it can fix it for us. So it's like a whole conversation. We concise. It's a whole conversation, so I can keep on going with Chachi Pitt as long as it's the same project, as long as I want and as much as I want, as long as I have the plus. If I don't, then as I mentioned before, everything's going to take longer, and I'm not going to have the privilege of having a couple of images to generate for the same thing. Alright, so I've got the result. And now Chachi Pitt has decided that, you know, it's an L shaped window, but the original one, you know, it's kind of tricky, but the original one is cupboards. So what I will do is that I'm going to download this one. This is how we fix this. And I'm going to do edit and I'll mark it up. And here, I'm going to mark up the blue. And I'll save options, save as a copy. I'll pick the downloads, save here. Okay, just override it. That's fine because it's here already. And I will fresh, take this and tell Chat TPT is that the blue highlight should continue as a wall cupboard. The glass is only on the left. Please keep everything, but fix this i And now Chachi Pit will try to give us even more results. So this is unrealistic. This is more realistic because, you know, it's a window continuing. But now I want it only the window to be on the bright side. I don't want this part. So let's see if Chachi Pitt is smart enough to change it. But as I told you guys, this is how we can achieve replacing models. You know, the main course of this lecture is to teach you guys how to replace models rather than to make an image perfect. But, you know, I'm trying to give you perspectives on how we can make our image as perfect as possible. All right, it has finished, but it didn't fix our problem. So I'm going to download this one and try to that GPT, I'm going to upload this image again. Wait, wait. Let's do something. Let's do something. I'll download this one, the previous one. Let's say, I'm going to download this one. And I'm going to refresh. I'll double click on it, and I'm going to mark it with something that is more visible. So let's mark it with orange. So there is a contrast, and I'm going to do this. I'll save it. I'll overwrite this one. Refresh. I've got it here. I'm going to drag and drop this Wever the orange is. This is not a window. This is a wall. Or I can say the orange markup is not a window. It is a wall. This should fix the error. If it didn't fix the error, then I'm already happy with what I got replacing the furniture, giving the perspective for the client. So let's see. And as I mentioned before, guys, this is already achieved by me. So if I go to the original file, it's already achieved. It has fixed it for me the way I want it. And, of course, the more we add into the prompt, the more it will change stuff in our rendering. So it's completely achievable even from the first time. But Chachi PT is giving us, you know, all the other possibilities that could be there. So let's see if it can fix it for the last time testing on this image. Now it has closed the entire window. Thank you, Chat, TPT. This is not what we're looking for. But as I guys told you, this is a tutorial to show you how you can change furniture or specific elements. So now what I want to do next is that I'm going to take the original image and retype the prompt in a new chat so it can work because now we are far away from what we're trying to achieve. You know, here we have a middle table. And we want this kind of to stay the way it is, you know. So now the setup is different and now I have to deal with the orange wall. So usually what I do is that I do a new conversation, and I put everything there again, and I try to prompt it brand new so I can get the exact result that I'm looking for. 17. 16_Furniture to CAD: How can I get the CAD side view and front of view of this chair with a prompt through Chachi PT? Great. So I'll do a new conversation, and I'll drag and drop my image, and I'll take this prompt and paste it here. Extract front side and top two DCAD views of the hairs from the perspective image with a clean black line and correct proportions. But I want to point something out is that if we do the side and the front and side and top, then the top will be cropped. So it can take only two views at the same time. Hopefully, in the new upgrade of hachPT it will understand this. So I will remove the top view side, so front and side to decad and I'll see what ha TipiT will give me. And then after that, I'm going to prompt for the side and the top, so I get them all. And then after that, I can put them into Convertio and get cat drawings for it to be able to, you know, put it in an interior or fix it in a certain way inside my scene. So let's see what we will be getting from hatchiPT. Okay, so it has gave us a result, but this is not the result that we are looking for. So I will be creating a new conversation again. Or let's just go back to the conversation that we have. Okay, here, I'm going to tell it. I'm going to put the image again, and I will copy the prompt, but I will change some things in the prompt is that I will say extract front and side two DCAD views of the chair from the three D perspective image. With clean black lines and correct proportions, maintaining the details of the hair. So again, this is what we got, and this is not what we're looking for, but Chachi PT sometimes, you know, run into errors. So I'm going to refresh Chachi Pit and I'll do a new conversation, and I'm going to give it the chair, and I'll do extract from this three D perspective image of the chair, two D cad, front and side view to not through with clean lines, clean black lines and correct proportions. I'll do a twist. So you are a professional architecture and interior drafter, and you got to extract from this three D perspective image of the chair a two D cad Let's see if now by giving a title for the AI in Chat TPT, it could generate our two D from our three Ds. So let's see if we're going to get that now. Great. So now I've got the correct drawings, but it's not a full image. So the image is this is correct, but the image is cropped. Regenerate the views without changing the design. But this time, make the two ortho graphic views fit into one image. Okay, so now it's better. You know, it's exactly fitted the two views into our image. And now we can take this and put it in convirtio and get our two D cad drawings. I hope you guys benefit from this lecture, see you guys in the next one. 18. 17_Mood board extraction: Alright, so now I will do a new conversation, and what I want to do now is that I want to extract a moodboard from this rendering in this style. So I'm going to take these two, drag it drop, and I'll do this prompt saying extract furniture, moodboard from image. Extract furniture, moodboard, from image, such as the reference mode board provided. Let's see what we can get. Perfect. This is what we got. But this is the mood board because the first one was giving us a network error. So let's just refresh Chat ChiPT again. And see if I'm going to get the result now. If I'm not going to get the result because of loss in connection, then I will re ask for it again and see what we will get. Okay, so I've left it and it's taking forever, so I'm going to refresh this page again. I'm going to copy. Oh, it's gone. Okay. And, I'll copy this one because this is wrong. It has give me the exact same one here, and I will do Control V, and I will take these two again, put them here. Wait for them to upload. Push. Let's see if it's going to work now. It should work now because the first one was hanging because of Internet connectivity. Rated furniture moodboard, material, textures, furniture elements, accessories. Yeah, pretty much. So would you like visual layout, extract moodboards similar to reference? Yes. So there we go. Now we've got an accurate mood board for the images that I provided. It comes with different styles, and I can, you know, specify which moodboard do I want more? Do I want an accessory moodboard, a furniture mood board, or a materials moodboard? And accordingly, Chachi PT will provide me with that. Again, all the imperfections of the process I've recorded for you guys to know that we don't usually get result from the first instance. We get it after trying and, you know, teaching Chachi PT what we're trying to achieve. So here are different results. You know, I've got a result here, and also I have another result for the furniture here. So you can specify the style of the moodboard that you need and what type of moodboard do you need. I hope you guys benefit from this one, see you guys in the next. 19. 18_Render to sketch: How can we achieve this sketch from this image? In a really quick and short tutorial. So any image I provide, I can give the prompt of make this sketch a black pen architecture sketch with some notes written above, and there we go. It has generated the image that we are looking for, and then I can ask it, make it an artistic watercolor sketch, and it has generated the watercolor sketch. So as you guys can see, it's pretty simple, pretty straightforward, and the sky is the limit with what we can achieve with Chat GPT. I hope you guys benefit from this chapter, see you guys in the coming ones. 20. 19_Concept generation and image enhacement : Alright, so we're back to Korea again, and we want to talk about how we can generate a concept from any image that we want. So actually, this part is a prompt to image rather than its image to another product. So I can go to image here, and here are some examples of prompts that I generated, and the quality is amazing. So you guys can see here I've prompted for a modern apartment building that is infill architecture with large windows. And here I've prompted for generate a modern resort cottage with large windows that are on an island in the middle of the sea. And it's pretty decent and pretty like the visual fidelity is crazy, guys. So describe an image. So I want or generate a civic building that is governmental, modern with large windows in the middle of the city. For example, let's say this is the project that I've got, and I want to generate this. Now, it actually gave me some branders. So to give you guys a glimpse, I can do generate a newseum building Artistic Museum, building that is modern with large windows in the middle of the city. Let's see how does this one look like. And guys pay attention to this is that this is completely free. So no need for you guys to be looking for other websites to, you know, subscribe to them and be able to generate concepts. This one actually generates concepts, and also you can choose the aspect ratio that you're looking for to generate the concept and also the resolution and all. So you can get many, many, many options from this, not just writing a prompt and getting an image from it. All right, so it's loading, but it's taking a long time, so I would refresh. That's actually, is it? It is generated. So it had generated concepts for me. And now if I want to take it on a really exotic scenario, so I can copy and paste this prompt or put it here, but remove Oh, this is actually, okay, so I need to write it generate a modern resort, cottages that are on an island in the middle of the sea. On a green island in the middle of the sea, and I want to do three by two. And now this one is faster, so it's generating in real time. Let's check the images. Wow. This is really nice. As a concept, guys, if you guys have no idea what you're going to do with your project, I mean, you can generate a couple of things to get approval from your client or your boss and you have a direction to go through. So are you looking for modern? Are you looking for contemporary or are you looking for somehow a modern but a traditional type of architecture? So you can name it all. You can even have like an island look and you can ask for an aerial shot or front shot or an isometric shot. So it's unlimited to what you guys can do. And for the image enhancers, as we mentioned before, you can enhance any images, any videos, you can edit in real time. And I want to show you take you guys through the history of the main things that I've done, and you guys can see how rea can do you a huge favor editing your images here as well. So I've tried to, you know, edit. Some images, they do drastic changes and really beautiful ones. Some images are minimal changes, like, not really worth it to do. But as you guys can see here, you know, some minimal changes here as well, some minimal changes. But really cool ones. Really having all these tools can give us power to what we're trying to create. So here, for example, you know, the aerial view this is the previous one. This is the one that we've edited. So it gives you a really good capture on your final result and what you want to do with it. So here, for example, this is the previous one. This is the enhanced one. And let's keep on going. You can even, you know, enhance your face. Here are some other projects that I've done, and I was able to enhance. You know, sorry about that, but they're really confidential ones. Here, for example, a sample. Of what you can do with your own image. So rea usually is a really good destination to stop by and, you know, explore what you can do. You can do a lot. You can do image generation, you can do video. You can do real time. You can do enhancer, can edit images, even video lip sync. This is new. You can train Crea. You can do three D objects, and you can do assets. So I hope you guys have a glimpse of what is Crea and how we can enhance our images and even generate concepts for our images. This leads us to the final chapter where we're going to talk about how to remove watermarks and how to generate an image, a three D from an image. See you guys in the next coming lectures. 21. 20_Water Mark and Image to 3D model: Alright, guys, so this leads us to the last chapter of our course where we're going to cover image to three D, and we're going to cover watermark removal. So please note for the watermark removal, we don't advise you on any type of plagiarism or stealing images from Google, from websites. We advise you to buy the images professionally if you're willing to take them. But this part of the course is for the people that already have license, but they want to you know, remove watermarks. And this is an example. I've got this image, and I've got a watermark, and here is another image where I remove the watermark. And basically, I need to hover over AIE's website, and from here, I need to AI tools, and let's see is a watermark removal. Mmm mm. Okay, great. So I need to hover down and go start for free here. And then I need to get my image the one that has watermark uploaded here. It's processing. And Voila, we've got the image with no watermarks, guys. At least this can take us 5 minutes on Photoshop. And at the best case scenario on the worst case scenario, it could take us an hour on Photoshop. If the watermark was hard and coming on objects that are complex, then, I need to be giving up on the details of the image for us to be able to achieve this clean watermark removal. So again, we don't advise you to do this for items that are not purchased, but for items that you have guys license on, you can use this tool to easing up your life. 22. 21_Image to 3D model: And our last lecture of discourse is the image to three D. So basically, we will learn how to get this image and put it and get this three D in return. So, guys, one thing we need to know is that our images, preferably, they must have a white Alpha background so we can get an accurate model. And how do we actually do this is that we get rendering, and then we extract this bed from. So for instance, I'm going to go back and I'll go to flux and I'll go to interior angles, and I'll take this I'll go to Chat TPT and I'll do a new chat. And I'm going to drag drop this image, and I will ask for I need the bit only in a white Wther white background white Alpha background to be able to use it in an AI image 23d website. Let's see what we get. So it has cut our bed from our image. And usually this takes time, especially when we're trying to achieve this on Photoshop. And it has made it an PNG type of thing. So if we open it, this is what we get. So I assume that we can use this as well. It's not only the white ones. So let me download this image. And now I'm going to go to a website called three D three. This is it. And I should create an account signing up into this website, creating an account, and then it will give us like 300 credits, which is enough to do almost 30 models, more than enough to show you the capabilities of this website. So I'll do generate a three D model. And I'll do image 23d, upload an image. I'll take this one and open. And here I've got some parameters, so I can do a private, smart lo poly generate in parts or auto. And each of these requires a certain amount of tokens. So I'll do generate for now. I'll see what we get from the generation of a PNG that does not have a white background. So as we can see, it is generating the three D model from here. And if we have the plus or the premium, we can speed up the process to get this image much faster. So about 30 seconds, and I should get a three D model where I need more than 30 seconds to navigate to a freelancing website and look for a three D modeler, and assign him to a job. So you guys can imagine how much time this is, you know, short cutting on us. Okay, great. So here is our three D model. So we click on the icon here and we get our three D model in the viewport to check it out. And it has exactly modeled Tippet for us. Relatively good quality, not the best quality out there, but I mean, it does the job, right? And I want to explain a further feature here that we can also take, but not always. We can just, you know, play with it and see if we can get a good result is that we can generate textures for this bed. So I've hovered over texture, and I'll do generate texture, and now it will take a couple of seconds to generate the texture for us according to the image. So we can use either an Alpha PNG image or we can use an Alpha white background to generate the three D model. Either or works. And the first thing that it does is that it generates a three D model, and then we can ask for the texturing. But the problem here is that three D repo usually gives us the three D texture as a stitch texture. And this is not always the best scenario for three D representation softwares because there is many mistakes that could be happening and also many errors, and maybe the stitching texture is, is definitely not the best optimized texture, but for a really quick, really, really quick texturing, it does an amazing job. Now if I want this model, I can do Export and from export, I can specify the file name, so I'll do BD and I can choose from USD FBX OBJ STL, GLB, and TMF. I'll do FBX and I'll do Export. And now it's exporting. It's collecting all the data together. And now it's downloading. So here I've got the FBX model. It's already downloaded. I can do OBJ, for example, and Export, and it will also export for me. So I've got all the different types, international types of exports that I can generate from this website and use it in my three D software either was RevetT D SMAx, ketchup, blender, Maya, you name it. And here I've got a three D tripo and I can do extract, and when I do an extract, it's going to give me the bed, the materials, and the stitch texture. As I mentioned, it's going to look this way whenever I'm assigning it. So maybe this is not the best practice, but for a quick one, it's an amazing practice. I hope you guys benefit from this lecture, and this whole entire course, really, guys, I've given you more than 20 workflows for you to be using inside your project. That would save you thousands of hours. Was a pleasure, and I'm really looking forward to see you guys in my upcoming courses where we can be talking about so many advanced AI tools and even unreal engine courses and even three D modeling courses and texturing courses coming from an architecture background. All of these is a must to cover, and I'm really excited to cover for you guys. So if you have any questions, please write me in the comments, and I'll be more than happy to answer them. Otherwise, good luck with your journey, and it's been a pleasure again, see you guys in our upcoming courses.