目前只在Ubuntu20.04+rocm5.2上搭建成功,需要切换到root权限搭建,否则Real-ESRGAN会自动下载nvidia的驱动(可能需要添加到环境变量中),搭建链接:Ubuntu22.04安装AMD显卡驱动和pytorch_V2077的博客-CSDN博客,项目链接:GitHub - xinntao/Real-ESRGAN: Real-ESRGAN aims at developing Practical Algorithms for General Image/Video Restoration.
安装Real-ESRGAN
$ cd ~
$ git clone https://github.com/xinntao/Real-ESRGAN.git
$ cd Real-ESRGAN
# 安装 basicsr - https://github.com/xinntao/BasicSR
# 我们使用BasicSR来训练以及推断
$ pip install basicsr
# facexlib和gfpgan是用来增强人脸的
$ pip install facexlib
$ pip install gfpgan
$ pip install -r requirements.txt
$ sudo python3 setup.py develop
图片修复
sudo python3 inference_realesrgan.py -n RealESRGAN_x4plus -i inputs --face_enhance
视频修复
安装
sudo apt-get -y install ffmpeg && pip3 install ffmpeg
修复
# rmvb转mp4
$ ffmpeg -i [inputname].rmvb -c:v libx264 -strict -2 [outputname].mp4
$ sudo python3 inference_realesrgan_video.py -n RealESRGAN_x4plus -i inputs.mp4 --face_enhance
或者
$ ffmpeg -i input.mp4 -qscale:v 1 -qmin 1 -qmax 1 -vsync 0 ./inputs/frame%08d.jpg
$ sudo python3 inference_realesrgan.py -n RealESRGAN_x4plus -i inputs --face_enhance
$ ffmpeg -i ./results/frame%08d_out.jpg -i input.mp4 -map 0:v:0 -map 1:a:0 -c:a copy -c:v libx264 -r 23.98 -pix_fmt yuv420p output.mp4
Python 脚本的用法
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
模型下载: