In this issue, we recommend Waifu2x Extension GUI, a tool that uses machine learning to enlarge images/videos/GIFs and video frame interpolation (patching)
Waifu2x Extension GUI uses deep convolutional neural networks to performsuper-resolutionenlargement (i.e. enlargement and noise reduction) onimages,GIFs, andvideos, as well asframe interpolation (i.e. frame supplementation) on videos.
Main features
-
- Automated processing of multiple media: Supports processing of images while achieving automated processing of GIFs& APNG and Video
- Full image style support: built-in multiple algorithms, whether it’s anime animation or photos you take daily& Video recordings can be processed for clarity
- Video Frame Insertion (Supplementary Frame): After super-resolution of the video, automatic frame insertion is performed to improve video fluency
Video image analysis: By intelligently analyzing video images, video processing speed can be accelerated and video quality can be improved
Powerful compatibility: Built in multiple engines, almost compatible with all modern Windows PCs
- Flexible multithreading settings: While providing comprehensive settings for each engine, you can also freely adjust the number of threads to fully utilize all the performance of the PC
- Multi GPU support: All built-in engines can support multiple graphics cards, fully utilizing all the performance of your PC
- Numerous custom settings: You can achieve automated workflows by adjusting the numerous custom settings within the software
- Intelligent settings: This software can automatically adjust some software settings based on your PC hardware information and the information you want to process files
- Easy to use graphical user interface
Built in super-resolution algorithm:
Waifu2x / SRMD / RealSR / Real-ESRGAN/ Real-CUGAN / Anime4K / ACNet
Built in super-resolution engine:
Waifu2x-caffe / Waifu2x-converter / Waifu2x-ncnn-vulkan
SRMD-ncnn-vulkan / RealSR-ncnn-vulkan / Anime4KCPP / SRMD-CUDA
RealESRGAN-NCNN-Vulkan / Real-CUGAN
Built in frame insertion algorithm:
RIFE / CAIN / DAIN
Built in frame interpolation engine:
rife-ncnn-vulkan / cain-ncnn-vulkan / dain-ncnn-vulkan
Tested on AMD RX 550, NVIDIA GeForce RTX 3060, and Intel UHD 620
Example
Comparison (3D realism),Using the built-in RealESRGAN NCNN Vulkan engine:
Comparison (2D anime),Using the built-in Waifu2x NCNN Vulkan engine:
Original 2D anime image 480×300 (. jpg 93.2 KB):
After 8x amplification, 3-level denoising, and compression, 3840×2400 (. jpg 525 KB),using the built-in Waifu2x NCNN Vulkan engine:
Original GIF 500 x 372 (493 KB):
1000 x 744 (3.77 MB) after 2x amplification, 2-level denoising, and GIF optimization,using the built-in Waifu2x NCNN Vulkan engine:

Waifu2x Extension GUI is currently only licensed for personal use, and more content can be read on your own