We have built a few video-search system by now, using USearch and UForm for embedding. They are only 256 dims and you can concatenate a few from different parts of the video. Any chance it would help?
A Mozilla PeerTube instance to join Blender's [1] in setting a good example would be a good place to start. The tech exists, and there's a large number of people on the ActivityPub fediverse ready to see it. Any organization that currently uses YouTube could do a similar mirror.
It doesn't yet, but that's something that the platform will support soon. Right now the platform uses AI-powered search to let you search your videos for interesting parts to clip. You can search for objects, actions, emotions, and text in video.
In that realm, one example I've seen pop up a lot where some kind of video needs to be shown off is "Big Buck Bunny" [1], a project to build an open movie using blender.
Sorry to hijack this post, but if anyone is interested in content recognition, we built one of the largest reverse search engines for video/audio [0] and are always looking for skilled engineers to do more.
I think our scale makes things very interesting. We currently index over 20 hours of video every second and to date we've indexed over 11.5B videos.
Feel free to reach out to me directly at r@pex.com.
reply