SAM 3 & 3D - Segment anything in images & video. Reconstruct 3D objects.
Meta introduces SAM 3 and SAM 3D. SAM 3 segments and tracks objects in images and videos using text or visual prompts. SAM 3D reconstructs 3D objects and bodies from single images. Try them now in the new Segment Anything Playground.
Replies
Flowtica Scribe
Hi everyone!
Generative AI has moved so fast that we've almost forgotten the main use case of the previous AI era: visual recognition. But the core of recognition is segmentation.
What if images could be segmented more precisely than ever, even tracking complex subjects in dynamic videos? And what if AI could infer a 3D model from that 2D segmentation with surprising accuracy?
Meta just released SAM 3 & 3D to do exactly that.
SAM 3 allows for precise segmentation and tracking in video via text prompts, while SAM 3D can reconstruct 3D objects from a single image.
Okay, sounds cool, but what can it actually do? Meta has prepared a rich Playground to demonstrate this. You can instantly hide sensitive objects to protect privacy, add creative motion trails to make your videos pop, or apply spotlight effects to focus attention exactly where you want it.
These capabilities are being integrated into apps like Instagram and will hugely empower creators and digitize industries.