Memvid – Video-Based AI Memory
github.comthis is incredibly dumb it is literally standard FAISS semantic search except the text content is converted to QR codes and stored as frames in an mp4, insanely convoluted lol
this has to be a troll
sadly have to agree. I asked a question about this on the issue tracker.
https://github.com/Olow304/memvid/issues/39 if anyone wants to follow along.
Cisco has a lot invested into optimizing video and voice communications in layer 2 network protocols. Does compressing this data into videos mean these network optimizations can be leveraged to speed up the delivery?
Hardware accellerated H.264 encode and decode are becoming more common in CPUs. Does this hardware accelleration benefit the general-purpose data storage in Memvid?