Real-ESRGAN VS stable-diffusion

It's not exactly what you're after, as it's anime specific and you need to process the video yourself (eg disassemble to frames, run the upscaler, then assemble back to a movie file), but Real-ESRGAN is really good:

https://github.com/xinntao/Real-ESRGAN/

It's pretty brilliant for cleaning up very old, low resolution anime.

Just a note you can run upscaling on your home desktop with Real-ESRGAN:

https://github.com/xinntao/Real-ESRGAN

Maybe Real-ESRGAN is a good fit? Even tho it's a couple of years old

I'm glad you noticed! I upscaled the icon from the wiki using Real-ESRGAN's 4xplus anime model, then photoshopped out the text. Worked far better than waifu2x.

A: Try using High-res Fix and R-ESRGAN 4x+ Anime6B as upscaler

Maybe try this one https://github.com/xinntao/Real-ESRGAN it may work even better.

Screenshotted from the official music video, upscaled to 4k using a state of the art ML model.

Do you happen to remember which chapter has that "scene"? You could also try to enhance it yourself, I did it using Real-ESRGAN, which is really easy to use.

Which is a 1-click install of Stable Diffusion with an alternative web interface. You can choose a different approach but this one is pretty simple and I am new to this stuff.

an invisible watermarking of the outputs, to help viewers identify the images as machine-generated.

It sounds like you just want to set up Stable Diffusion to run locally. I don't think your computer's specs will be able to do it. You need a graphics card with a decent amount of VRAM. Stable diffusion is in Python as is almost every AI open source project I've seen. If you can get your hands on a system with an Nvidia RTX card with as much VRAM as possible, you're in business. I have an RTX 3060 with 12 gigs of VRAM and I can run stable diffusion and a whole variety of open source LLMs as well as other projects like face swap, Roop, tortoise TTS, sadtalker, etc...

Use specific GPU on multi GPU systems · Issue #87 · CompVis/stable-diffusion · GitHub

I dealt with this in one of my posts. At least SD 1.1 till 1.5 are all trained on a batch size of 2048. The version pretty much everyone uses (1.5) is first pretrained at a resolution of 256x256 for 237K steps on laion2B-en, at the end of those training steps it will have seen roughly 500M images in laion2B-en. After that it is pre-trained for 194K steps on laion-high-resolution at a resolution of 512x512, which is a subset of 170M images from laion5B. Finally it is trained for 1.110K steps on LAION aesthetic v2 5+. This is easily verified by taking a glance at the model card of SD 1.5. Though that one doesn't specify for part of the training exactly which aesthetic set was used for part of the training, for that you have to look at the CompVis github repo. Thus at the end of it all both the most recent images and the majority of images will have come from LAION aesthetic v2 5+ (seeing every image approx 4 times). Realistically a lot of the weights obtained from pretraining on 2B will have been lost, and only provided a good starting point for the weights.

stable diffusion · CompVis/stable-diffusion@2ff270f · GitHub

Here for example --> https://github.com/CompVis/stable-diffusion