BREAKING 🚨 Meta AI launches DINOv3, a open source SOTA-enabling vision foundation model, trained with pure self-supervised learning (SSL) at scale. - High quality dense features with semantic and geometric scene understanding - No more reliance on annotated data - power many applications: MLLMs, video&3D understanding, robotics, gen. models - SOTA on all 3 vision tasks: detection, segmentation and depth Meta FAIR team continues to deliver gems in AI.
