StereoPolicy learns robot 3D perception from stereo pairs, hits 59% success

TheZachMueller · x · 2026-09-10

Lambda's StereoPolicy gives robots 3D perception without depth sensors, depth maps, point clouds, or LiDAR — learning 3D structure directly from stereo image pairs. Across 5 real-world tabletop tasks it achieved 59% success vs. 42% for RGB, 41% for RGB-D, and 14% for PointNet.

Original post →

More from Embodied

Embodied channel →