7 ms·
Meta Segment Anything Model 3
- trevorhlynn 10mo agoThis was front page for a while last week https://news.ycombinator.com/item?id=45982073 https://news.ycombinator.com/item?id=45982073
- stronglikedan 10mo agowhat is old is new again
- dang 10mo agoThanks! Macroexpanded: Meta Segment Anything Model 3 - https://news.ycombinator.com/item?id=45982073 https://news.ycombinator.com/item?id=45982073 - Nov 2025 (133 comments) p.s. This was lobbed onto the frontpage by the second-chance pool (https://news.ycombinator.com/item?id=26998308 https://news.ycombinator.com/item?id=26998308) and I need to make sure we don't end up with duplicate threads that way.
- Workaccount2 10mo agoI do a test on multimodal LLMs where I show them a dog with 5 legs, and ask them to count how many legs the dog has. So far none of them can do it. They all say "4 legs". Segment anything however was able to segment all 5 dog legs when prompted to. Which means that meta is doing something else under the hood here, and may lend itself to a very powerful future LLM. Right now some of the biggest complaints people have with LLMs stems from their incompetence processing visual data. Maybe meta is onto something here.
- jampekka 10mo agoSegmentation doesn't need to count legs. I'd guess something like YOLO could segment 5 legged dogs too.
- chompychop 10mo agoYOLO is not a segmentation model.
- jampekka 10mo agohttps://docs.ultralytics.com/tasks/segment/ https://docs.ultralytics.com/tasks/segment/
- lucasban 10mo agoI thought it was a joke about YAML
- chompychop 10mo agoThanks! TIL there's a class of segmentation models with the YOLO naming scheme.
- Der_Einzige 10mo agoLol you obviously haven't seen what cheats for FPS games look like in the last 3 years. https://github.com/Babyhamsta/Aimmy https://github.com/Babyhamsta/Aimmy
- PunchTornado 10mo agoI doubt that gemini 3 cannot do it.
- nerdsniper 10mo agoYou don’t need segmentation to count legs. Object detection can do that. DeepLabCut from 2020 perhaps.
- the_duke 10mo agoSide question: what are the current top goto open models for image captioning and building image embeddings dbs, with somewhat reasonable hardware requirements?
- NitpickLawyer 10mo agoTry any of the qwen3-vl models. They have 8, 4 and 2B models in this family.
- Glemkloksdjf 10mo agoI would suggest YOLO. Depending on your domain, you might also finetune these models. Its relativly easy as they are not big LLMs but either image classification or bounding boxes. I would recommend bounding boxes.
- smallerize 10mo agoWhich YOLO?
- Glemkloksdjf 10mo agoAny current one. they are easy to use and you can just benchmark them yourself. I'm using small and medum. Also the code for using it is very short and easy to use. You can also use ChatGPT to generate small exepriments to see what fits your case better
- throwaway314155 10mo agoThere aren’t any YOLO models for captioning and the other models aren’t robust enough to make for good embedding models.
- Glemkloksdjf 10mo agoYou can get labels out of the classifier and bounding box models. They are super fast. Its just an alternative i'm mentioning. I would assume a person knowing a little bit of that domain. Otherwise the first option would be CLIP i assume. llm-vl is just super slow and compute intensive.
- vessenes 10mo agoReleased 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 under 15 seconds.
- nl 10mo agohttps://ai.meta.com/blog/sam-3d/ https://ai.meta.com/blog/sam-3d/ for those interested.
- Fraterkes 10mo agoAre 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
- seanw265 10mo agoI 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 https://github.com/facebookresearch/MHR
- daemonologist 10mo agoFor 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).
- vessenes 10mo agoI’ve only used the playground. But I think they are actual meshes - they don’t have any of the weird splat noise at the edge of the objects, and they do not seem to show similar lighting artifacts to a typical splat rendering.
- phkahler 10mo agoWhich (if any) of these models could run on a RaspberryPi for object recognition at several FPS?
- enoch2090 10mo agoSurprisingly, SAM3 works bad on engineering drawings while SAM2 kinda works, and VLMs like Qwen3-VL works as well
- zubiaur 10mo agoHad good luck with Gemini 2.5, SAM3 failed miserably with PIDs.
- retinaros 10mo agoyeah I tried too. Im trying a fine tuning on PIDs.
- enoch2090 10mo agoLooking forward to your progress! Just checked the paper and it says the underlying backbone is still DETR. My guess would be that SAM3 uses more video frames during the training process and caused the dilution of sparse engineering-paper-like data.
- shashanoid 10mo agoMiss the old segment anything page, used it a lot. This UI I found very complex to use
- bradyriddle 10mo agoSame. Checkout https://github.com/MiscellaneousStuff/meta-sam-demo 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.
- stronglikedan 10mo ago> I'm surprised it's not getting more attention. Remember, it's not the idea, it's the marketing!
- colkassad 10mo agoBeen waiting days to get approval to download this from huggingface. What's up with that?
- knicholes 10mo agoI was approved within about 10 minutes for "Segment Anything 3"
- observationist 10mo agoAlternative 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.
- colkassad 10mo agoThanks, got it and it's working wonders for my use case.
- tschellenbach 10mo agosame here, didn't get approval
- cheesecompiler 10mo agoThis would be convenient for post-production and editing of video, e.g. to aid colour grading in Davinci Resolve. Currently a lot of manual labour goes into tracking and hand-masking in grading.
- Will-Reppeto 10mo ago[dead]
- aliljet 10mo agoI wonder how effective this is medical scenarios? Segmenting organs and tumors in cat scans or MRIs?
- maelito 10mo agoI 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.
- vanjoe 10mo agoFor a long time I've wanted to use something like this to remove advertisements from hockey games.The moving ads on the boards are really annoying. Maybe I'll get around to actually doing that one of these days.