Everyone records videos with their smartphones, whether it’s concerts, kids, or stuff you think you’ll upload to TikTok. Truth is though, you might go back and watch those moments in your camera roll, but you’ll probably never edit them and make them into something shareable. Video editing takes a long time. Interesting things are happening with AI computer vision these days.

Ilias Haddad created an app that will index and semantically describe your videos. He made it to solve a specific problem he had:

I had 2,207 GoPro videos, and I need to rewatch them to find interesting moments from my cycling journey. I built a project to index them locally on my M1 Max using open-source ML models, search for those moments, and send the best clips straight to my DaVinci Resolve timeline.

NJ Singh built something similar:

Every AI video editor on the market assumes your footage is already labeled. Mine is IMG_*.mov and DJI_*.mp4 across folders with names like Mara june 2024 backup final FINAL. Eddie [an AI video editor,] can search by transcript, but none of these tools can find “the elephant on the hill at golden hour” against an unlabeled archive.

The AI editor is solving the wrong problem. Or more precisely, it’s solving the second problem; the first problem is the index.