meisa233

1. Install nv-codec-headers

   git clone https://github.com/FFmpeg/nv-codec-headers.git
   cd nv-codec-headers
   # Check the lowest version of nvidia driver in README
   # If you need to switch to ohter branch
   git branch -a
   git checkout remotes/origin/sdk/11.1 
   make
   sudo make install PREFIX=~/ffmpeg_build/

meisa233

本次文档测试的系统为Ubuntu 16.04，很多通过apt install安装的库都很破旧
想要vlc实现解码稀有编码格式必须能够完整地安装vlc，通过snap或apt安装的vlc没有办法以root用户运行，因此只能重新编译vlc，并在编译前配置时加入允许root用户运行的选项(enable-run-as-root)，当然如果转码程序本身可以在不用root用户运行的情况下就能够正常运行的话直接通过snap install vlc就可以（通过apt安装的vlc十分破旧），不需要再按照这个文档编译vlc

1. 安装依赖
   （Install Requirements）

sudo apt-get install build-essential pkg-config libfreetype6-dev libfontconfig1-dev libx11-dev libxext-dev libxinerama-dev \
libv4l-dev libxcb-shm0-dev libxcb-xv0-dev libav-tools libavcodec-dev \
libavformat-dev zoo libswscale-dev libqt4-dev autopoint flex gettext bison \ 
liblua5.2-dev lua5.2 luajit libxcb-composite0-dev libxcb-xkb-dev

meisa233

This answer is from Chatgpt 3.5. The information about error:

qt.qpa.plugin: Could not find the Qt platform plugin "xcb" in ""

Answer:

export QT_PLUGIN_PATH=/path/to/Qt/plugins
E.g. export QT_PLUGIN_PATH=/media/sdtv/新加卷/MyCode/MiVOS/venv/lib/python3.7/site-packages/PyQt5/Qt5/plugins

meisa233

1.update apt

ChangeAptSource

sudo apt update
sudo apt upgrade

Install dependencies

sudo apt-get install zlib1g-dev libbz2-dev libssl-dev libncurses5-dev libsqlite3-dev libreadline-dev tk-dev libgdbm-dev libdb-dev libpcap-dev xz-utils libexpat1-dev liblzma-dev libffi-dev libc6-dev

meisa233

1. Install dependency packages

sudo apt update
sudo apt install build-essential cmake pkg-config
sudo apt-get install libavcodec-dev libavformat-dev libswscale-dev libv4l-dev
sudo apt-get install libxvidcore-dev libx264-dev
sudo apt-get install libatlas-base-dev gfortran
sudo apt-get install libjpeg8-dev libtiff5-dev libjasper-dev libpng12-dev
sudo apt-get install libgtk-3-dev
sudo apt-get install libatlas-base-dev gfortran

meisa233

extract fixed number frames

ffmpeg -ss 00:00:00 -i input.mp4 -vsync 0 -an -vframes 600 -f image2 -start_number 0 %05d.png

start time: 00:00:00
frame number: 600
output format: %05d.png
start_number: start number of output images
extract frames by specified duration

meisa233

add this line after import torch

torch.backends.cudnn.enabled = False

meisa233

1. Bringing Old Films Back to Life CVPR 2022
   Paper: https://arxiv.org/abs/2203.17276
   Code: https://github.com/raywzy/Bringing-Old-Films-Back-to-Life
   

2. ReBotNet: Fast Real-time Video Enhancement CVPR 2022
   Paper: https://arxiv.org/abs/2303.13504
   Project Website: https://jeya-maria-jose.github.io/rebotnet-web/
   

meisa233

SOME FEATURES NOT AVAILABLE.
Python version:3.8.10

1. Install Requirements

sudo apt-get install python3-dev python3-pip python3-setuptools \
                     build-essential cmake libatlas-base-dev \
                     libavcodec-dev libavformat-dev libavutil-dev \
                     libswscale-dev libjpeg-dev libpng-dev libtiff-dev \
                     libv4l-dev libxvidcore-dev libx264-dev \
                     libopenblas-dev liblapack-dev libhdf5-dev \

meisa233

1. Compile davs2

git clone https://github.com/pkuvcl/davs2.git
cd davs2/build/linux
./configure --prefix=/home/sdtv/ffmpeg_build --enable-pic --enable-shared
make -j8
make install

2. Compile xavs2