Bardienus Duisterhof (@BDuisterhof)
2025-06-06 | โค๏ธ 322 | ๐ 57
Imagine if robots could fill in the blanks in cluttered scenes.
โจ Enter RaySt3R: a single masked RGB-D image in, complete 3D out. It infers depth, object masks, and confidence for novel views, and merges the predictions into a single point cloud. https://rayst3r.github.io/ https://x.com/BDuisterhof/status/1930986607655649647/video/1
๋ฏธ๋์ด
