Search papers, labs, and topics across Lattice.
Affiliation:
2
0
4
A self-supervised approach reveals that a shared canonical object frame can emerge from noisy video data, achieving pose estimation accuracy without manual annotations.
