pablogod/video2x

ALL INFO https://github.com/ZurMaD/video2x

This page covers the usage of Video2X's Docker image.

Prerequisites

Only Nvidia GPUs are supported at this time. The host must have driver version 440, uniform with which that is inside of the Docker container.

The host must have package nvidia-container-toolkit installed. The variable distribution is simply our Linux distribution + version ID, such as debian10. If your distribution is not "supported" (e.g., Kali), you can change it manually.

shell
# Add the package repositories
distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add -
curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list

sudo apt-get update && sudo apt-get install -y nvidia-container-toolkit nvidia-container-runtime
sudo systemctl restart docker

Creating and Running Docker Container

Below is an example of how you can use Video2X with Docker. This command upscales the video sample_input.mp4 two times with Waifu2X NCNN Vulkan and outputs the upscaled video to sample_output.mp4. The Docker image will be pulled automatically during this process if it's not pulled already.

shell
docker run --rm -it --gpus all -v /dev/dri:/dev/dri -v $PWD:/host pablogod/video2x:4.6.0 -d waifu2x_ncnn_vulkan -r 2 -i sample_input.mp4 -o sample_output.mp4

In detail, each of the components of the command above does the following.

shell
docker run --rm \ # run the Docker and delete it when finished
       -it \ # run in interactive mode with a pseudo TTY
       --gpus all \ # pass all host GPUs into Docker
       -v /dev/dri:/dev/dri \ # mount host direct rendering infrastructure into Docker for hardware acceleration
       -v $PWD:/host \ # mount host's current working directory into Docker's /host directory
       pablogod/video2x:4.6.0 \ # use image pablogod/video2x version 4.6.0 (pull if necessary)
       -d waifu2x_ncnn_vulkan \ # use waifu2x-ncnn-vulkan for upscaling
       -r 2 # upscaling ratio
       -i sample_input.mp4 \ # specify input file path
       -o sample_output.mp4 \ # specify output file path

Manually Pulling Image

In case you would like to pull the image manually or update the local image, you can pull the image from Docker Hub just like you would any other images. Replace TAG with the tag of the latest version (e.g., 4.6.0). Do not pull the latest tag since it might be unstable.

shell
docker pull pablogod/video2x:TAG