Meta’s new AI-powered Segment Anything Model 2 (SAM 2) is capable of identifying which pixels belong to a target object in an image or video. This means that it can segment any object and consistently follow it across all frames of a video in real-time, whether it be for video editing purposes or mixed reality experiences.
SAM 2’s ultra fast inference capabilities could inspire new ways of selecting and interacting with objects in real time, even during livestreams. This model could also be used to aid research in science and medicine, with two such examples being tracking endangered animals in drone footage or localizing regions in a laparoscopic camera feed during surgery. Try it out now.
- Dive into extraordinary experiences with a mixed reality headset that transforms your home into an exciting new playground, where virtual elements...
- It’s the most powerful Quest yet*, featuring next-level performance with more than double the graphic processing power of Quest 2. (*Based on the...
- Experience more immersion with dazzling visuals with the 4K+ Infinite Display (a nearly 30% leap in resolution from Quest 2) and rich 3D audio with...
SAM 2 could also be used to track a target object in a video to aid in faster annotation of visual data for training computer vision systems, including the ones used in autonomous vehicles. It could also enable creative ways of selecting and interacting with objects in real-time or in live videos,” said Meta.
