https://github.com/cunnyplapper/CuNNy

Convolutional upscaling Neural Network, yeah!
https://github.com/cunnyplapper/CuNNy

anime cunny magpie mpv super-resolution uoh upscaling

Last synced: 2 months ago
JSON representation

Convolutional upscaling Neural Network, yeah!

• Host: GitHub
• URL: https://github.com/cunnyplapper/CuNNy
• Owner: cunnyplapper
• License: lgpl-3.0
• Created: 2024-02-26T20:45:06.000Z (4 months ago)
• Default Branch: master
• Last Pushed: 2024-04-21T16:41:34.000Z (2 months ago)
• Last Synced: 2024-04-21T22:07:06.545Z (2 months ago)
• Topics: anime, cunny, magpie, mpv, super-resolution, uoh, upscaling
• Language: HLSL
• Homepage:
• Size: 43.8 MB
• Stars: 24
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Lists

• awesome-mpv - CuNNy - Cute and funny CNN-based upscaler optimized for anime. (Other)

README

        
# CuNNy - Convolutional upscaling Neural Network, yeah!

Nice, small, and fast realtime CNN-based upscaler.

Currently very new and immature 😭.

Supports exporting to an mpv ~~meme~~shader!

And now a Magpie effect!

# Usage

mpv shaders are found inside the `mpv/` directory.

Non-denoising variants are found inside the `results/` directory.

Magpie effects are found inside the `magpie/` directory.

The order of best quality -> worst quality is sorted by the 2nd number first,

then the first number. So `16x16` > `4x16` > `16x8`.

Conversely the order of fastest -> slowest would be the reverse, with `0x4`

being the fastest and `3x32` being the slowest.

Variants:

- `DS`: Trained to denoise & sharpen images.

- `NVL`: Trained on VN screenshots/CG.

There are versions of the mpv shaders use 8-bit `dp4a` instructions. They can be

many times faster than the standard upscaling shader depending on if your

hardware supports accelerated `dp4a` instructions. Requires `vo=gpu-next` with

`gpu-api=vulkan`. They can be found inside the `dp4a/` subdirectories.

# Training

Tested training with PyTorch nightly. If any errors arise try using nightly.

Prepare data by running `sh scripts/build.sh`, then `sh scripts/split.sh

 `, then `py scripts/proc.py <128-grids> `.

To train `py train.py   ` where `N` is the number of internal

convolutions and `D` is the number of feature layers.

Convert the resulting model to an mpv shader by running

`py mpv.py `.

Convert the resulting model to a Magpie effect by running

`py magpie.py `.

Trains very fast on my machine.

# Quality

See `results/`.

# License

LGPL v3