ソ瑠璃(soruly)
ソ瑠璃(soruly)
Yes the new API response has the full URL, so you don't need to construct the url youself.
I would like to talk about a "video scene cutter" I've been using for generating video preview for [trace.moe 's telegram bot](http://t.me/whatanimebot) https://github.com/soruly/trace.moe-media
This search engine is not a neural network trained by machine learning, it's just a perceptual image hash database on large amonut of anime. The screenshots you submit are not...
I also have plans to detect anime series by reading video files. However I can't decide how many thumbnails should be taken from one video for search would yield accurate...
In the long run I'd prefer to make whatanime.ga distributed, so anyone with skills would be able to setup an image database of their own.
With my new improvements made on searching, I think a decent quad core CPU would be able to handle any search in 5 seconds. And for hashing video, a 24-minute...
I've opensourced the distributed indexing system, you can take a look https://github.com/soruly/sola
whatanime.ga has moved to https://trace.moe https://www.patreon.com/posts/moving-to-new-22212117
The current system requires 128GB RAM and ~200GB nvme SSD storage to run. I don't know if it's possbile to run on hugging face.
Let's get back to this when hugging face supports docker-compose