Telegram Server:
https://t.me/joinchat/KTRznBIPPNCbHkUqnwT8pA
Dandere2x makes upscaling videos with waifu2x much faster by applying compression techniques. In other words, Dandere2x uses techniques used from video streaming (such as Youtube or Netflix) to decrease the time needed to upscale a video using Waifu2x.
- Block Matching Video Compression
- Interpolated Frame Video Compression
- Minimum / maximum quality loss (set by JPEG standards)
- Color fading (fade to black)
- Custom block sizes
- Custom scaling sizes (1x, 2x, 3x)
- Interactive GUI
I wrote a quick wiki-entry to answer the question. I hope it's accessible to most.
https://github.com/aka-katto/dandere2x/wiki/How-Dandere2x-Works
I made a brief video on how to use dandere2x. Dandere2x is constantly changing, so this video may be outdated.
Dandere2x is in development and in research. Much of my current thoughts can be found on my subreddit or on twitter. I highly encourage anyone who needs help / has questions to either submit an issue on GitHub or ask other users on the subreddit.
https://www.reddit.com/r/Dandere2x/
Please do! If you don't understand any parts of my code, I would be happy to explain/re-organize my code for you. Furthermore, if you can't code, I can give you a list of tasks you can do to help further dandere2x development!
Q: What is the difference between Dandere2x and Video2x?
A: Video2x aims for quality over speed. If you're looking for the best-possible looking video, Video2x is more streamlined and provides lossless upscaling. Dandere2x, on the other hand, is still in development, is (can) much faster than video2x, but at the cost of being lossy.
Q: How does Dandere2x work?
A: https://github.com/aka-katto/dandere2x/wiki/How-Dandere2x-Works
Video2x: A lossless video enlarger/video upscaler achieved with waifu2x.
This project relies on the following software and projects.
- waifu2x-caffe
- waifu2x
- FFmpeg
- STB Image
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent