moonbox-ai/Real-ESRGAN-MLX

Real-ESRGAN MLX

MLX builds of Real-ESRGAN, made by Moonbox for Apple Silicon Macs. Each precision is a folder; keep a folder's files together. Only the parts listed below are converted: the rest of the pipeline comes from the original repository unless the section below says otherwise.

Files

The last column is how far each build is from the original PyTorch model on the same inputs (see the section below for what was measured).

Precision	File	Size	sha256	Checked against the original
bf16	bf16/config.json	0.2 kB	6317d8073fe7	PSNR 40.8 dB mean / 30.7 min, LPIPS <= 0.015
bf16	bf16/mlx_model.py	4.8 kB	dca0f44b9087	PSNR 40.8 dB mean / 30.7 min, LPIPS <= 0.015
bf16	bf16/model.safetensors	33.5 MB	c21e6b4f63ff	PSNR 40.8 dB mean / 30.7 min, LPIPS <= 0.015
fp16	fp16/config.json	0.2 kB	a5848755fc43	PSNR 54.3 dB mean / 46.9 min, LPIPS <= 0.0004
fp16	fp16/mlx_model.py	4.8 kB	dca0f44b9087	PSNR 54.3 dB mean / 46.9 min, LPIPS <= 0.0004
fp16	fp16/model.safetensors	33.5 MB	5173c577ace0	PSNR 54.3 dB mean / 46.9 min, LPIPS <= 0.0004

How to run

Real-ESRGAN x4plus (RealESRGAN_x4plus.pth, the general-photo model) ported to MLX. Each folder holds model.safetensors, config.json and mlx_model.py, the port (only mlx and numpy needed):

import numpy as np, mlx_model
from PIL import Image
model = mlx_model.load("fp16")
out = mlx_model.upscale(model, np.array(Image.open("in.png").convert("RGB")))   # 4x, tiled (256 px tiles)
Image.fromarray(out).save("out.png")

Which one: fp16/ is the closest to the original (54 dB PSNR); bf16/ is visually the same (LPIPS <= 0.015) but drifts on some images (30.7 dB worst). There are no 8bit / 6bit / 4bit builds: the network is all convolutions, which MLX doesn't quantize, and at 33 MB it gains nothing from it.

Precision	Size	PSNR vs PyTorch fp32 (mean / worst)	LPIPS (worst)	Time, ~130x130 -> 4x on an M5
fp16/	33.5 MB	54.3 / 46.9 dB	0.0004	0.25 s
bf16/	33.5 MB	40.8 / 30.7 dB	0.015	0.28 s

How these were made

From Real-ESRGAN at revision a4abfb2979: the original weights loaded in MLX and stored as bf16 or fp16, or quantized with MLX's affine quantization (8bit, 6bit, 4bit, group size 64; layers MLX can't quantize, such as convolutions and norms, stay bf16). Every build was checked against the original before upload; the sha256 column above is what the files must hash to.

License

BSD-3-Clause. These files are a derivative of the original model and carry its license: it is copied in this repository, and it governs your use of these files. The original model card follows unchanged.

---

Original model card

English | 简体中文

👀Demos | 🚩Updates | ⚡Usage | 🏰Model Zoo | 🔧Install | 💻Train | ❓FAQ | 🎨Contribution

🔥 AnimeVideo-v3 model (动漫视频小模型). Please see [anime video models] and [comparisons]
🔥 RealESRGAN_x4plus_anime_6B for anime images (动漫插图模型). Please see [anime_model]

1. :boom: Update online Replicate demo:
2. Online Colab demo for Real-ESRGAN: | Online Colab demo for for Real-ESRGAN (anime videos):
3. Portable Windows / Linux / MacOS executable files for Intel/AMD/Nvidia GPU. You can find more information here. The ncnn implementation is in Real-ESRGAN-ncnn-vulkan

Real-ESRGAN aims at developing Practical Algorithms for General Image/Video Restoration.
We extend the powerful ESRGAN to a practical restoration application (namely, Real-ESRGAN), which is trained with pure synthetic data.

🌌 Thanks for your valuable feedbacks/suggestions. All the feedbacks are updated in feedback.md.

---

If Real-ESRGAN is helpful, please help to ⭐ this repo or recommend it to your friends 😊
Other recommended projects:
▶️ GFPGAN: A practical algorithm for real-world face restoration
▶️ BasicSR: An open-source image and video restoration toolbox
▶️ facexlib: A collection that provides useful face-relation functions.
▶️ HandyView: A PyQt5-based image viewer that is handy for view and comparison
▶️ HandyFigure: Open source of paper figures

---

📖 Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data

[Paper]   [YouTube Video]   [B站讲解]   [Poster]   [PPT slides]
Xintao Wang, Liangbin Xie, Chao Dong, Ying Shan
Tencent ARC Lab; Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences

---

🚩 Updates

• ✅ Add the realesr-general-x4v3 model - a tiny small model for general scenes. It also supports the -dn option to balance the noise (avoiding over-smooth results). -dn is short for denoising strength.
• ✅ Update the RealESRGAN AnimeVideo-v3 model. Please see anime video models and comparisons for more details.
• ✅ Add small models for anime videos. More details are in anime video models.
• ✅ Add the ncnn implementation Real-ESRGAN-ncnn-vulkan.
• ✅ Add RealESRGAN_x4plus_anime_6B.pth, which is optimized for anime images with much smaller model size. More details and comparisons with waifu2x are in anime_model.md
• ✅ Support finetuning on your own data or paired data (i.e., finetuning ESRGAN). See here
• ✅ Integrate GFPGAN to support face enhancement.
• ✅ Integrated to Huggingface Spaces with Gradio. See Gradio Web Demo. Thanks @AK391
• ✅ Support arbitrary scale with --outscale (It actually further resizes outputs with LANCZOS4). Add RealESRGAN_x2plus.pth model.
• ✅ The inference code supports: 1) tile options; 2) images with alpha channel; 3) gray images; 4) 16-bit images.
• ✅ The training codes have been released. A detailed guide can be found in Training.md.

---

👀 Demos Videos

Bilibili

• 大闹天宫片段
• Anime dance cut 动漫魔性舞蹈
• 海贼王片段

YouTube

🔧 Dependencies and Installation

• Python >= 3.7 (Recommend to use Anaconda or Miniconda)
• PyTorch >= 1.7

Installation

1. Clone repo

   git clone https://github.com/xinntao/Real-ESRGAN.git
   cd Real-ESRGAN
   

2. Install dependent packages

   # Install basicsr - https://github.com/xinntao/BasicSR
   # We use BasicSR for both training and inference
   pip install basicsr
   # facexlib and gfpgan are for face enhancement
   pip install facexlib
   pip install gfpgan
   pip install -r requirements.txt
   python setup.py develop
   

---

⚡ Quick Inference

There are usually three ways to inference Real-ESRGAN.

1. Online inference
2. Portable executable files (NCNN)
3. Python script

Online inference

1. You can try in our website: ARC Demo (now only support RealESRGAN_x4plus_anime_6B)
2. Colab Demo for Real-ESRGAN | Colab Demo for Real-ESRGAN (anime videos).

Portable executable files (NCNN)

You can download Windows / Linux / MacOS executable files for Intel/AMD/Nvidia GPU.

This executable file is portable and includes all the binaries and models required. No CUDA or PyTorch environment is needed.

You can simply run the following command (the Windows example, more information is in the README.md of each executable files):

./realesrgan-ncnn-vulkan.exe -i input.jpg -o output.png -n model_name

We have provided five models:

1. realesrgan-x4plus (default)
2. realesrnet-x4plus
3. realesrgan-x4plus-anime (optimized for anime images, small model size)
4. realesr-animevideov3 (animation video)

You can use the -n argument for other models, for example, ./realesrgan-ncnn-vulkan.exe -i input.jpg -o output.png -n realesrnet-x4plus

Usage of portable executable files

1. Please refer to Real-ESRGAN-ncnn-vulkan for more details.
2. Note that it does not support all the functions (such as outscale) as the python script inference_realesrgan.py.

Usage: realesrgan-ncnn-vulkan.exe -i infile -o outfile [options]...

  -h                   show this help
  -i input-path        input image path (jpg/png/webp) or directory
  -o output-path       output image path (jpg/png/webp) or directory
  -s scale             upscale ratio (can be 2, 3, 4. default=4)
  -t tile-size         tile size (>=32/0=auto, default=0) can be 0,0,0 for multi-gpu
  -m model-path        folder path to the pre-trained models. default=models
  -n model-name        model name (default=realesr-animevideov3, can be realesr-animevideov3 | realesrgan-x4plus | realesrgan-x4plus-anime | realesrnet-x4plus)
  -g gpu-id            gpu device to use (default=auto) can be 0,1,2 for multi-gpu
  -j load:proc:save    thread count for load/proc/save (default=1:2:2) can be 1:2,2,2:2 for multi-gpu
  -x                   enable tta mode"
  -f format            output image format (jpg/png/webp, default=ext/png)
  -v                   verbose output

Note that it may introduce block inconsistency (and also generate slightly different results from the PyTorch implementation), because this executable file first crops the input image into several tiles, and then processes them separately, finally stitches together.

Python script

Usage of python script

1. You can use X4 model for arbitrary output size with the argument outscale. The program will further perform cheap resize operation after the Real-ESRGAN output.

Usage: python inference_realesrgan.py -n RealESRGAN_x4plus -i infile -o outfile [options]...

A common command: python inference_realesrgan.py -n RealESRGAN_x4plus -i infile --outscale 3.5 --face_enhance

  -h                   show this help
  -i --input           Input image or folder. Default: inputs
  -o --output          Output folder. Default: results
  -n --model_name      Model name. Default: RealESRGAN_x4plus
  -s, --outscale       The final upsampling scale of the image. Default: 4
  --suffix             Suffix of the restored image. Default: out
  -t, --tile           Tile size, 0 for no tile during testing. Default: 0
  --face_enhance       Whether to use GFPGAN to enhance face. Default: False
  --fp32               Use fp32 precision during inference. Default: fp16 (half precision).
  --ext                Image extension. Options: auto | jpg | png, auto means using the same extension as inputs. Default: auto

Inference general images

Download pre-trained models: RealESRGAN_x4plus.pth

wget https://github.com/xinntao/Real-ESRGAN/releases/download/v0.1.0/RealESRGAN_x4plus.pth -P weights

Inference!

python inference_realesrgan.py -n RealESRGAN_x4plus -i inputs --face_enhance

Results are in the results folder

Inference anime images

Pre-trained models: RealESRGAN_x4plus_anime_6B
More details and comparisons with waifu2x are in anime_model.md

# download model
wget https://github.com/xinntao/Real-ESRGAN/releases/download/v0.2.2.4/RealESRGAN_x4plus_anime_6B.pth -P weights
# inference
python inference_realesrgan.py -n RealESRGAN_x4plus_anime_6B -i inputs

Results are in the results folder

---

BibTeX

@InProceedings{wang2021realesrgan,
    author    = {Xintao Wang and Liangbin Xie and Chao Dong and Ying Shan},
    title     = {Real-ESRGAN: Training Real-World Blind Super-Resolution with Pure Synthetic Data},
    booktitle = {International Conference on Computer Vision Workshops (ICCVW)},
    date      = {2021}
}

📧 Contact

If you have any question, please email xintao.wang@outlook.com or xintaowang@tencent.com.

🧩 Projects that use Real-ESRGAN

If you develop/use Real-ESRGAN in your projects, welcome to let me know.

• NCNN-Android: RealSR-NCNN-Android by tumuyan
• VapourSynth: vs-realesrgan by HolyWu
• NCNN: Real-ESRGAN-ncnn-vulkan

    GUI

• Waifu2x-Extension-GUI by AaronFeng753
• Squirrel-RIFE by Justin62628
• Real-GUI by scifx
• Real-ESRGAN_GUI by net2cn
• Real-ESRGAN-EGUI by WGzeyu
• anime_upscaler by shangar21
• Upscayl by Nayam Amarshe and TGS963

🤗 Acknowledgement

Thanks for all the contributors.

• AK391: Integrate RealESRGAN to Huggingface Spaces with Gradio. See Gradio Web Demo.
• Asiimoviet: Translate the README.md to Chinese (中文).
• 2ji3150: Thanks for the detailed and valuable feedbacks/suggestions.
• Jared-02: Translate the Training.md to Chinese (中文).