Depth Anything 3

Depth Anything 3 is a single transformer model developed by researchers for recovering visual depth and geometry from any viewpoint, distinguishing itself by unifying multiple 3D vision tasks into one efficient framework.

Two-hop subgraph: this entity, every entity it directly relates to, and every entity those neighbors relate to. Drag a node, scroll to zoom, click to inspect — or click any neighbor and re-center the atlas there.