Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models

news.ycombinator.com

Post reply on HN
Print view View original

31–40 of 129 posts

Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models

#31

Post by teovall »

iliashad wrote: post #25
robrain wrote: post #23

DaVinci 21 has indexing built-in (AI IntelliSearch). Not to diminish the work you did, but this is now available to many users (probably only Studio users since it has AI in the name)

Yes, I didn’t look at it. But does it upload your videos to the cloud or process them locally? And does it allow to provide custom faces data to help labeling faces in your videos ? I think Adobe premiere pro have it as well but cloud processed

The AI features in DaVinci Resolve are all processed locally. It does not currently have face tagging.

Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models

#32

Post by iliashad »

teovall wrote: post #31
iliashad wrote: post #25

Earlier quoted context omitted.

Yes, I didn’t look at it. But does it upload your videos to the cloud or process them locally? And does it allow to provide custom faces data to help labeling faces in your videos ? I think Adobe premiere pro have it as well but cloud processed

The AI features in DaVinci Resolve are all processed locally. It does not currently have face tagging.

That’s great to know, thank you!

Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models

#35

Post by robrain »

teovall wrote: post #31
iliashad wrote: post #25

Earlier quoted context omitted.

Yes, I didn’t look at it. But does it upload your videos to the cloud or process them locally? And does it allow to provide custom faces data to help labeling faces in your videos ? I think Adobe premiere pro have it as well but cloud processed

The AI features in DaVinci Resolve are all processed locally. It does not currently have face tagging.

Haven’t tried it yet, and I don’t know if it matches OP’s requirements, but the blurb says “You can even search for individual faces”

https://www.blackmagicdesign.com/products/davinciresolve/wha...

Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models

#36

Post by pduggishetti »

vorticalbox wrote: post #20
pduggishetti wrote: post #5

Earlier quoted context omitted.

You'll need a lora for this, porn content rejection is heavy. Or you'll need a abliterated model, not sure if vision also works. You might want to add something like yolo finetune to detect scenes + face recognition too.

Vision still works perfectly fine in abliterated models.

Never tried any of this for porn, just speaking out how I would go about it tbh!

Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models

#38

Post by tontonius »

if anyone is interested in searching large video collections local and offline I suggest taking a look at Jumper https://docs.getjumper.io

comes with some nifty features like NLE- integrations, people search, MCP, API etc

Disclaimer: one of the co-founders

Re: I indexed 669 GB of my GoPro videos using my M1 Max computer and local ML models

#40

Post by ingvay7 »

iliashad wrote: post #30
m3kw9 wrote: post #17

Grab frames, lower res, classify, combine meta data. Write to sql

Not really. Grab frames, lower res, classify, combine metadata, transcribe the audio, convert those data (text, visual and audio) to embedding, save them over a vector DB and SQL DB. Which helped me to do semantic search, RAG, search using a screenshot of the video to find the exact the moment in the video plus search using an audio file as well. And other features unlocked with vector DB

Really cool work and workflow. strongly prefer this kind of local, open pipeline that i control over a dependency on Adobe tools and lock ins.
Post reply on HN