[0:00] Generative AI has opened the door to
[0:02] allowing anyone to create incredible
[0:04] images just using a computer and a few
[0:07] text prompts. Now, the platform that
[0:08] gets the least amount of attention is
[0:10] Stable Diffusion, but it might just be
[0:12] the hidden gem you're looking for.
[0:14] Unlike other major generative AI
[0:16] platforms like Adobe's Firefly or
[0:18] OpenAI's Dali or even Midjourney, Stable
[0:20] Diffusion is completely open source,
[0:23] which in short is free. So, I'd want
[0:25] Padme to wear her iconic [music] white
[0:27] turtle attire. See that image is coming
[0:29] in. It's looking great. There's a few
[0:31] issues going on in the lower parts of
[0:32] the frame. In this particular section,
[0:34] I'm changing how the body structure
[0:36] looks. Again, let's increase that to
[0:38] four images. Okay, we're going to pause
[0:40] it right there. Remember, there are no
[0:42] limitations in the software, which means
[0:44] the responsibility is on you.
[0:49] Stable diffusion lets you run on your
[0:51] own GPU instead of on the cloud, which
[0:53] allows you to tweak every minute setting
[0:56] that you can think of. Now, if that's
[0:57] not enough to convince you to learn this
[0:59] software, it's also the only platform
[1:01] with zero limitations, but it also comes
[1:04] with its own drawbacks. It's tough to
[1:06] get started with, but that's where we
[1:08] come in. Today, we're going to help you
[1:09] with everything from getting started,
[1:11] installation to even more complicated
[1:13] things like perfecting your prompts and
[1:15] getting the best possible image. Now,
[1:17] today's episode is a little bit
[1:19] different. It's just you, me, and a
[1:21] computer. And we're going to get through
[1:22] this together. It's also the first time
[1:24] that we haven't had someone behind the
[1:26] camera, so expect things to go wrong.
[1:30] First off, what is stable diffusion? In
[1:32] short, it's an image generation model
[1:34] created by stable AI. It turns text
[1:37] prompts into images. What makes stable
[1:40] diffusion really special is that it's
[1:42] open source, which in short means that
[1:44] it's free, but it also means that you
[1:46] can download it, bring it into your own
[1:48] computer, and customize it to your own
[1:50] art style.
[1:53] Now, in a majority of this video, we
[1:54] will work with a software application
[1:56] called Fucus. This is a front-end
[1:59] interface that allows us to interact
[2:01] with the stable diffusion model. Now,
[2:03] don't let any of those words scare you
[2:05] off. Think of stable diffusion as the
[2:07] engine, the code that's doing all the AI
[2:09] image processing. Focus is like the car
[2:12] that's built around the engine. It's
[2:14] what has all the buttons and the
[2:16] controls and the interface that allows
[2:18] you to interact with the engine without
[2:20] really messing around with the tech.
[2:22] Okay, enough talking. Let's jump into
[2:24] the application. The first thing you
[2:25] want to do is go to the GitHub
[2:26] repository for Focus. We'll leave a link
[2:28] in the description. After you download
[2:30] the file, you want to extract the file
[2:31] and you'll see and run.bat. You want to
[2:33] click on the run.bat file and start the
[2:36] installation process. Remember, the
[2:37] download file will be around 50 GB. So,
[2:40] make sure you have that capacity before
[2:42] you get started. By default, you'll get
[2:44] the standard model, but you'll also get
[2:45] the option to run the animate and
[2:47] realistic models as well. For right now,
[2:49] we're not going to change any of the
[2:50] default behavior of the software, which
[2:52] means as it launches, it'll look for new
[2:54] models, keep the software updated, which
[2:56] is generally what you want. But if you
[2:58] want to change that, you can use these
[3:00] two command lines. Let's do realistic
[3:02] right now. And now we just let it do its
[3:03] thing. Another quick tip that I like to
[3:05] use is keeping task manager open. As
[3:08] long as I see activity on the GPU, I
[3:10] know that the software is working with
[3:12] the GPU to get it collected.
[3:15] Awesome. And we are finally ready to
[3:16] start generating images. You'll notice a
[3:18] few things. The image generation window
[3:20] at the top, the text prompt window at
[3:21] the bottom where we can type things like
[3:23] cats on a window lid. And at the bottom,
[3:25] you can see input image. We'll go into
[3:27] that in detail. It's something that I
[3:29] use in great detail. Pain enhance, which
[3:31] we won't talk about too much.
[3:32] essentially post-processing steps that
[3:34] you can take on your final image to
[3:36] increase the resolution and detail. As
[3:38] we hit generate, you can see that there
[3:40] is an immediate spike in your GPU. So,
[3:42] you'll notice the image comes in a
[3:44] little fuzzy, then it cleans up over
[3:46] time. Okay, with those images in place,
[3:47] we can look at them in detail by
[3:49] clicking on each individual image. The
[3:50] second image is a really good example of
[3:52] the kind of problems that this kind of
[3:54] model has. You can see that the eye on
[3:55] the left is perfect, but the eye on the
[3:57] right just doesn't have the detail we
[3:58] need. So, there's two things that we can
[4:00] think about at this point. We can click
[4:01] on advanced and we can first look at the
[4:03] presets. We are using the realistic
[4:05] model which is what generates such fine
[4:07] detail but there's lots of other
[4:08] variations that you can try out here.
[4:10] There's also performance depending on if
[4:12] you want to take a qualitative or
[4:13] quantitative approach. Right now we're
[4:16] prioritizing speed of delivery, but we
[4:18] can also change that to quality to
[4:20] generate higher quality images. Now a
[4:22] feature that I use all the time is
[4:24] changing the number of images, but for
[4:26] right now let's do four images.
[4:29] >> [music]
[4:31] >> That's pretty incredible. The first
[4:32] three images have come in. I love the
[4:34] detail here. There's lots of detail in
[4:36] the hair and in the eyes. Both eyes look
[4:38] perfect. The second image is not quite
[4:40] as good. It has really great
[4:41] perspective. Lots of detail in the
[4:43] bricks and the window, but not quite as
[4:45] much detail on the cat itself. You can
[4:47] see that the eye on the right isn't
[4:48] perfect. Okay, this is great perspective
[4:50] cuz there's a beautiful window. It's a
[4:52] low angle shot. Again, not quite as much
[4:54] detail on the cat itself. And the final
[4:56] image has also come through. Oh, this is
[4:57] amazing. And you can see the cat looking
[4:59] through the window. Lots of detailing on
[5:01] the glass and the light and the texture
[5:03] on the skin. We're going to build a
[5:04] crazy Star Wars poster no one has ever
[5:06] seen of Padme fighting off Anakin
[5:09] Skywalker. But first, let's build a
[5:11] cyberpunk city with crazy [music]
[5:12] detail.
[5:14] So, you can notice the kind of detail
[5:16] that I'm putting in here. I'm I'm
[5:17] writing out very specifically what I
[5:19] want to see. So, a cyberpunk city at
[5:21] night. So, I'm specifying what the
[5:23] lighting looks like. Glowing neon
[5:25] lights. someone have that look where
[5:26] there's lots of light from the buildings
[5:27] themselves. And I'm specifically calling
[5:29] for cinematic lighting. With AI models,
[5:32] the more specific details you can give,
[5:34] the more likely you are to get a
[5:35] successful final image. Okay, so those
[5:38] two images have come through. They're
[5:39] both fairly similar, which is something
[5:40] I don't really like. Ideally, I want
[5:42] variation on shot. Also, something I
[5:44] like to do is watch the model as the
[5:46] image generates just to see if it's
[5:48] something that's in alignment with what
[5:50] I want. If it's completely off, I can
[5:51] choose to skip through that image. Okay.
[5:54] And that's pretty great. Everything's
[5:55] generally looking okay, but nothing is
[5:57] specifically looking good. That's also
[5:59] another problem with the AI generation
[6:00] model for wide and detailed shots. It's
[6:03] harder to get fine details correct when
[6:05] there's lots of details spread across
[6:07] the image. Let's say we love this image
[6:08] in general, but we want to enhance
[6:10] certain details. With that in mind,
[6:12] let's refine our model some more. Okay,
[6:14] now we've got two new images. Let's look
[6:16] at both of them. I love the detailing on
[6:17] this. Generally, everything seems okay.
[6:19] There's lots of lost details in the
[6:21] buildings in the background, but I think
[6:22] I can generally fix those as we go. The
[6:24] second image is a lot softer and there
[6:27] is a lot of billboards and written
[6:29] detail which [music] I probably won't be
[6:31] able to fix. What if we have a great
[6:32] image, but we want to refine that image.
[6:34] That's where image input comes in.
[6:36] First, if you want to take this image
[6:38] and we just want to scale it up, bring
[6:39] in new resolution, we can use image
[6:41] upscale. So, you just drag that image in
[6:43] here. First, we look at the bottom row.
[6:45] Upscaling by 1.5, upscaling by 2x. that
[6:48] essentially just expands the resolution,
[6:51] maintaining as much of the image as
[6:52] possible. And then you have upscaling
[6:54] fast 2x, which is the same thing, but it
[6:56] processes it a little faster with a
[6:58] little less accuracy. So, first let's do
[6:59] a 1.5 upscale. Now, as that image comes
[7:02] in, we can see we already have a lot
[7:04] more detail. Okay, the first image is
[7:06] in. Let's take a look at what that looks
[7:07] like. So, you're already seeing so much
[7:09] more detail in the building in the
[7:10] foreground. All the problems of the
[7:12] background have now been resolved. All
[7:14] the lines are straight. All the windows
[7:15] are visible. Generally, everything's
[7:17] looking good. There is still a little
[7:19] bit of specific problems I'm seeing,
[7:21] especially on this billboard on the
[7:22] right, parking lot down below, but the
[7:24] cars don't look perfect. We can work on
[7:26] those specifically. Okay, let's look at
[7:28] this image with some detail. That's not
[7:30] going to work cuz it's so close to the
[7:32] foreground element, which is this
[7:33] building. So, to refine this, we're
[7:34] going to use impaint. This allows you to
[7:36] refine and work on specific details of
[7:38] your image with great detail. There's a
[7:40] brush, and anything you brush over will
[7:42] then be changed. Anything that's outside
[7:45] the brush radius will not be affected.
[7:47] So, we've got three options here.
[7:49] Impaint, improve, and modify. Impaint is
[7:51] a great technique if you want to change
[7:53] something of your image with subtle
[7:55] variation and keep the general look of
[7:56] the image the same. The first thing
[7:58] you'll notice is that the GPU is using
[7:59] all of its processing power on just that
[8:01] one zone of the image, which means
[8:03] you'll get a lot of resolution in just
[8:05] that one area. Has a little motel
[8:07] looking building with a swimming pool in
[8:09] front. It's lots of details in terms of
[8:11] cars and people, which may not be what
[8:13] we want because it's going to attract a
[8:14] lot of attention. And then the second is
[8:18] a black building with a few windows. You
[8:20] can see that I can actually drag the
[8:22] image from my image generation window
[8:24] into my impaint window. This is a pretty
[8:27] common technique of refining your image
[8:29] as you move it back and forth within the
[8:31] software itself. And this is a good time
[8:33] to talk about the other two impend
[8:35] features as well. Improve detail is used
[8:37] very often. This is when you want to
[8:39] increase the resolution of something in
[8:41] the background, right? We can see far
[8:42] more detail in that image. You can see
[8:44] the specific floors of the building. You
[8:47] can see through some of the windows. You
[8:48] can see some trees and shbery brought in
[8:50] front of the building. Really quickly,
[8:52] let's look at modify content as well.
[8:54] This is a powerful tool when you want to
[8:55] make a dramatic shift of your image and
[8:58] then later go back and refine it using
[9:01] the refineer tool. So, let's look at one
[9:02] of these. And that's what that car path
[9:04] looks like. Again, here you can see a
[9:05] lot of imperfections in the car, in the
[9:07] floor, in the building next to the car
[9:08] park. All of which will need to be
[9:10] refined if you want to use it in your
[9:12] final image. But for right now, we're
[9:14] just going to revert back to the image
[9:15] that we had and look through our final
[9:17] settings. At this point, you probably
[9:19] get the idea of image generation, but
[9:21] let's dive into something a little bit
[9:22] more obscure that uses more advanced
[9:25] features of the software.
[9:29] So, [music] let's think of something
[9:30] that we can't find on the internet.
[9:32] though. Anakin Skywalker and Padme
[9:35] lightsaber dual Star Wars franchise high
[9:37] detail cinematic lighting at night.
[9:39] [music] All right, let's see what this
[9:41] looks like. Again, let's increase that
[9:42] to four images. Okay, we're going to
[9:45] pause it right there. So, this is where
[9:47] you need to be careful as you generate
[9:49] images. You need to make sure that you
[9:50] don't generate any images that's not
[9:52] safe for work. And this is a good time
[9:54] as any to talk about the power of these
[9:56] creative tools. Remember, there are no
[9:58] limitations in these software, which
[10:00] means the responsibility is on you. Make
[10:02] sure you don't create any content that's
[10:03] harmful, misleading, or violates
[10:06] privacy. Another tool that we can use to
[10:07] safeguard our content is negative
[10:09] prompts. Now, these are things that you
[10:11] don't want in your image. So, by
[10:13] default, I'm getting things like
[10:14] unrealistic, saturated, big nose,
[10:16] painting, drawing, sketch. These are all
[10:19] prompts that have come in default by the
[10:21] software. But now, I'm going to also
[10:22] include not safe for work as a prompt.
[10:25] And you can expand on that list. Okay.
[10:27] And those images have come through. They
[10:29] both have their own challenges. So over
[10:31] here, both Anakin and Padme are sharing
[10:33] a lightsaber, which doesn't really make
[10:34] the most sense. In the second image, I
[10:36] kind of have their hands crossed over.
[10:38] Both which we can fix. And we can do
[10:40] that by bringing this image, dragging it
[10:42] in using impaint and correcting for. But
[10:45] we're not going to do that right now
[10:46] because there is a holistic problem in
[10:48] the image. And that's the fact that I
[10:50] don't like the perspective that we're
[10:52] getting. Ideally, I'd want to see
[10:53] something similar to the Star Wars
[10:54] poster. It's a low angle shot with lava
[10:57] in the background and each character
[10:58] fighting aggressively against each
[11:00] other. Now I could give this in the form
[11:03] of a prompt, but I can also use image
[11:06] prompt. Image prompt essentially allows
[11:08] you to input an image and generate a
[11:11] similar image. Now, but for right now,
[11:13] let's turn off all of our advanced
[11:15] features. You can go over to image
[11:17] prompt and you can drag your image in
[11:19] and you have a few features that will
[11:20] show up. Now, if you don't have this bar
[11:22] at the bottom, you just want to scroll
[11:23] down to the bottom and click advanced.
[11:26] That'll give you four options. Now,
[11:27] image prompt will generally scan the
[11:29] image very similar to a textbased
[11:31] prompt. It'll attempt to understand
[11:33] what's happening in the image and it'll
[11:34] generally use that as a suggestion for
[11:36] generation. You can see that it's
[11:37] generally taking the idea of the image
[11:39] and generating something similar. But in
[11:41] this case, what we really need is
[11:43] something that's visually similar. So,
[11:45] we have two real options. Pyrochni and
[11:48] CPDS. Pyrochani essentially maps the
[11:50] position of characters and takes it into
[11:53] your new image which would be ideal for
[11:56] this particular image and CPDS
[11:58] essentially uses contrast, color and
[12:00] saturation to generate a similar image.
[12:03] So in this particular case, Pyrocheni
[12:05] would work perfect. Okay, we've got four
[12:07] images in. Let's let the remaining
[12:08] images come in. Let's look at the first
[12:10] one. So this first image, the first
[12:11] thing you'll notice is that it's fairly
[12:13] low resolution. There's not a lot of
[12:14] detail in the face structure. The
[12:16] lightsaber's broken. I don't like how
[12:18] the sand looks relative to the mountain.
[12:20] The second image is much better. And
[12:21] then it's converted Obi-Wan into some
[12:24] version of Anakin. So, that's really
[12:26] great in terms of perspective, in terms
[12:27] of each lightsaber. Lightsabers are red,
[12:30] but I'm assuming I can make those
[12:31] changes. Third image. I don't like it as
[12:34] much. The resolution is not quite there.
[12:36] The perspective isn't great. Anakin's
[12:38] looking a lot bigger than Padme. It's
[12:40] probably one we're going to avoid. Image
[12:41] four. That's looking much better. I love
[12:44] the perspective of this. This next image
[12:46] is completely broken. perspectives are
[12:48] wrong. Seems to be a dual lightsaber
[12:49] which Padme is holding from the wrong
[12:51] side. So, we're not going to use that
[12:53] image. Final image looks like two Padme
[12:55] fighting each other. Again, it's less of
[12:57] a problem because I know I can change
[12:59] one of those characters, [music] but I'm
[13:00] not a big fan of the perspective. I'm
[13:02] not a big fan of the background. So,
[13:04] with all of that in mind, I think I'm
[13:06] going to go with this image as the image
[13:08] I'm going to refine to final.
[13:11] Okay. Okay. So, the first thing I'll do
[13:12] is drag that image into impaint and
[13:15] start erase my previous painting and
[13:17] start working on specific sections of
[13:20] the image. So, I'd want Padme to wear
[13:22] her iconic white battle attire. So,
[13:24] these images are already more in
[13:26] direction of where it needs to be. This
[13:28] time, I'm only going to change up the
[13:29] outfit itself. Just highlighting very
[13:31] specifically. So, our first section of
[13:33] images are coming in and they're all
[13:34] looking pretty great. So, I love the
[13:36] position here. Doesn't look perfect in
[13:38] terms of outfit. Second outfit looks
[13:40] much better, but there's lots of
[13:41] problems with the hand structure. The
[13:42] hand seems a lot smaller than it needs
[13:44] to be. The foot structure isn't great.
[13:46] This third image is better in terms of
[13:48] hand structure, but there is a broken
[13:49] limb right here, and there's a few
[13:51] issues going on in the lower parts of
[13:53] the frame. So, here's a pro tip. In this
[13:54] particular section, I'm changing how the
[13:56] body structure looks, but I don't really
[13:59] want to change the general position. I
[14:01] find that it's helpful to highlight only
[14:03] parts of the image. So you'll notice
[14:04] here I've left the foot open and the
[14:07] back of the ankle open as well along
[14:08] with the waist of the character. That
[14:09] way I can change just the middle section
[14:11] of the image without affecting the
[14:12] overall look of the Now I'm liking the
[14:14] overall structure here. I just want to
[14:16] fix this broken limb. I'd like to change
[14:18] the shoes the character is wearing.
[14:19] Maybe reduce some of this armor. And
[14:21] that's what I'm going to do now. Now you
[14:22] can see that after a certain point I'm
[14:24] going to start running parallel
[14:25] processes. In this section I've
[14:27] highlighted Padme's hair very
[14:28] specifically. So I'm trying to get her
[14:30] hair tied up and it's proving to be a
[14:32] little bit difficult. But the important
[14:33] point here is that you need to keep a
[14:35] reference ready. I'm trying to match
[14:37] those references. Having a reference
[14:39] will always help get the details right
[14:41] and anchor your character in reality.
[14:43] So, one thing we'll definitely need to
[14:45] do is work on character features,
[14:48] especially in the face. You want to have
[14:49] as much detail as possible in the face.
[14:51] So, even if things aren't perfect, most
[14:54] people won't notice. And you can see I'm
[14:57] already painting in the second image
[14:58] while that first image is generated. I'm
[15:00] going to need to work on that next. So,
[15:02] Anakin, Star Wars, detailed face, it's a
[15:04] lot better. We're not going to perfect
[15:07] anything. We're going to do the best we
[15:08] can and really just go through all of
[15:10] the different tools and functions. If
[15:12] you'd want to see us explore that in
[15:14] another video where we deep dive into
[15:16] creating hyper realistic images, let us
[15:18] know in the comments below. Now, in this
[15:20] next section, I want to show you how to
[15:21] expand an image. Now, you've probably
[15:22] seen something similar if you've used
[15:24] Adobe's Photoshop with content of
[15:26] airfield, but it's quite powerful in
[15:28] focus because you can really control and
[15:30] refine how the expansion work prompt you
[15:32] put in here is very specifically the
[15:34] information that you want in the
[15:35] background and in the expanded areas.
[15:38] But ideally, I'd like to have this in
[15:40] landscape format. So, what I'm going to
[15:41] do is I'm going to take that image, put
[15:43] it back into content expansion, and just
[15:46] generate the left and right side of the
[15:48] image. Okay, that's looking really
[15:50] great. I love how that looks. Now, I
[15:51] don't really like how the characters are
[15:53] standing. I'm seeing a lot of
[15:55] imperfections in the hand structure,
[15:57] mannequin's leg structure, but generally
[15:59] this is a great starting base
[16:00] considering we generated this image in
[16:02] under an hour. That's a really great
[16:04] starting point. Now, there's a lot more
[16:05] you can do here using the refiner tools.
[16:07] You can bring in things like smoke and
[16:09] fog. You can work on the character
[16:10] outfits, bring in detail, so things like
[16:12] the shoes and the hair. I'll also look
[16:14] back at my references and really
[16:16] understand where I missed the ball. And
[16:17] that brings us to the end of this
[16:19] episode. If you like this video, hit the
[16:21] like and subscribe buttons. Now, there
[16:23] is a lot more we can do to perfect this
[16:25] photo.