Video2X

Video2X runs AI video upscaling and frame interpolation on your own hardware. It's for people who want to increase a video's resolution or generate extra frames for smoother motion, including those working with anime footage. The project is open source under AGPL-3.0.

Its choice of processing methods is a useful distinction. Anime4K, Real-ESRGAN and Real-CUGAN handle upscaling, while RIFE handles frame interpolation. Video2X also accepts custom MPV-compatible GLSL shaders, so you're not limited to its bundled processing methods. The neural models run through ncnn with Vulkan GPU acceleration.

Windows users get a graphical interface, and the tool also runs on Linux. Docker and Podman container images provide another route on Linux and macOS. Local processing uses your machine's hardware; the optional Google Colab notebook runs the workload on Google's servers for people who don't have a suitable GPU of their own.

Hardware compatibility matters here: Video2X requires a Vulkan-capable GPU, with support spanning compatible NVIDIA, AMD and Intel hardware. Its precompiled binaries also require an AVX2-capable CPU. This is a GPU-based video processing tool, so those requirements are worth checking before choosing it.

The C/C++ processing pipeline avoids using extra disk space for intermediate files. You only need storage for the final output.