What I've Learned in the Past Year Spent Building an AI Video Editor

Re:What I've Learned in the Past Year Spent Building an AI V... by liotier on Tuesday September 24, 2024 @08:12AM (#41635675)

Derushing in general is the most time consuming, so not only language pattern recognition but also image recognition: "From the rushes, extract all the sequences with bicycle crashes to give me a pile of clips to use in my edit" !

[ Parent | on HN ]

Re:What I've Learned in the Past Year Spent Building an AI V... by burningion on Tuesday September 24, 2024 @08:28AM (#41635779)

Yes, agreed.

I film a bunch of skateboarding, and it can take tens of tries to land a trick. Similarly, there's usually an unique sound that signals a trick was finally landed.

Good multi-modal search and discovery is a huge part of cracking the editing problem.

[ Parent | on HN ]

Re:What I've Learned in the Past Year Spent Building an AI V... by liotier on Tuesday September 24, 2024 @09:04AM (#41636055)

Looks like https://kino.ai addresses that derushing stage, but as a specialized tool rather than as a function inside a video editor - which makes a lot of sense to me.

[ Parent | on HN ]

Re:What I've Learned in the Past Year Spent Building an AI V... by sitkack on Tuesday September 24, 2024 @09:51AM (#41636451)

Detect the cheer everyone makes when the trick lands. Lots of proxy indicators to key off of.

[ Parent | on HN ]

Re:What I've Learned in the Past Year Spent Building an AI V... by trinix912 on Tuesday September 24, 2024 @01:57PM (#41639199)

Tens? It sometimes takes my crew hundreds of tries (all on DV tapes).

How far have you been able to come with search for trick variations? It would be interesting to see a system that can reliably recognize what’s switch, nollie vs fakie etc. Then have it generate a list of all tricks for each skater and perhaps outstanding fails. Just some thoughts.

[ Parent | on HN ]