Meta Segment Anything Model 3 • Hacker News Board

Re: Meta Segment Anything Model 3

#31

Post by seanw265 » Tue, Nov 25, 2025, 5:12 PM UTC

Fraterkes wrote: post #25

vessenes wrote: post #14

Released last week. Looks like all the weights are now out and published. Don’t sleep on the SAM 3D series — it’s seriously impressive. They have a human pose model which actually rigs and keeps multiple humans in a scene with objects, all from one 2D photo (!), and their straight object 3D model is by far the best I’ve played with - it got a really very good lamp with translucency and woven gems in usable shape in u…

Are those the actual wireframes they're showing in the demos on that page? As in, do the produced models have "normal" topology? Or are they still just kinda blobby with a ton of polygons

I haven’t tried it myself, but if you’re asking specifically about the human models, the article says they’re not generating raw meshes from scratch. They extract the skeleton, shape, and pose from the input and feed that into their HMR system [0], which is a parametric human model with clean topology.

So the human results should have a clean mesh. But that’s separate from whatever pipeline they use for non-human objects.

[0]: https://github.com/facebookresearch/MHR

Re: Meta Segment Anything Model 3

#32

Post by maelito » Tue, Nov 25, 2025, 5:12 PM UTC

I wonder if this can be used to track an object's speed. E.g. a vehicle on a road. It would need to recognize shapes, e.g. car model or average size of a bike, to guess a speed.

Re: Meta Segment Anything Model 3

#33

Post by knicholes » Tue, Nov 25, 2025, 5:13 PM UTC

colkassad wrote: post #23

Been waiting days to get approval to download this from huggingface. What's up with that?

I was approved within about 10 minutes for "Segment Anything 3"

Re: Meta Segment Anything Model 3

#34

Post by stronglikedan » Tue, Nov 25, 2025, 5:17 PM UTC

trevorhlynn wrote: post #2

This was front page for a while last week https://news.ycombinator.com/item?id=45982073

what is old is new again

Re: Meta Segment Anything Model 3

#35

Post by stronglikedan » Tue, Nov 25, 2025, 5:19 PM UTC

bradyriddle wrote: post #29

shashanoid wrote: post #22

Miss the old segment anything page, used it a lot. This UI I found very complex to use

Same. Checkout https://github.com/MiscellaneousStuff/meta-sam-demo It's a rip of the previous sam playground. I use it for a bunch of things. Sam 3 is incredible. I'm surprised it's not getting more attention.

> I'm surprised it's not getting more attention.

Remember, it's not the idea, it's the marketing!

Re: Meta Segment Anything Model 3

#36

Post by observationist » Tue, Nov 25, 2025, 5:29 PM UTC

colkassad wrote: post #23

Been waiting days to get approval to download this from huggingface. What's up with that?

Alternative downloads exist. You can find torrents, and match checksums against the HF downloads, but there are also mirrors and clones right there in HF which you can download without even having to log in.

Re: Meta Segment Anything Model 3

#37

Post by tschellenbach » Tue, Nov 25, 2025, 5:40 PM UTC

colkassad wrote: post #23

Been waiting days to get approval to download this from huggingface. What's up with that?

same here, didn't get approval

Re: Meta Segment Anything Model 3

#38

Post by visioninmyblood » Tue, Nov 25, 2025, 5:47 PM UTC

vessenes wrote: post #14

Released last week. Looks like all the weights are now out and published. Don’t sleep on the SAM 3D series — it’s seriously impressive. They have a human pose model which actually rigs and keeps multiple humans in a scene with objects, all from one 2D photo (!), and their straight object 3D model is by far the best I’ve played with - it got a really very good lamp with translucency and woven gems in usable shape in u…

you can download them at https://github.com/facebookresearch/sam3. for 3d https://github.com/facebookresearch/sam-3d-objects

Re: Meta Segment Anything Model 3

#39

Post by daemonologist » Tue, Nov 25, 2025, 5:58 PM UTC

Fraterkes wrote: post #25

vessenes wrote: post #14

Released last week. Looks like all the weights are now out and published. Don’t sleep on the SAM 3D series — it’s seriously impressive. They have a human pose model which actually rigs and keeps multiple humans in a scene with objects, all from one 2D photo (!), and their straight object 3D model is by far the best I’ve played with - it got a really very good lamp with translucency and woven gems in usable shape in u…

Are those the actual wireframes they're showing in the demos on that page? As in, do the produced models have "normal" topology? Or are they still just kinda blobby with a ton of polygons

For the objects I believe they're displaying Gaussian splats in the demo, but the model itself can also produce a proper mesh. The human poses are meshes (it's posing and adjusting a pre-defined parametric model).

Re: Meta Segment Anything Model 3

#40

Post by daemonologist » Tue, Nov 25, 2025, 6:03 PM UTC

the_duke wrote: post #7

Side question: what are the current top goto open models for image captioning and building image embeddings dbs, with somewhat reasonable hardware requirements?

For pure image embedding, I find DINOv3 to be quite good. For multimodal embedding, maybe RzenEmbed. For captioning I would use a regular multimodal LLM, Qwen 3 or Gemma 3 or something, if your compute budget allows.