Search papers, labs, and topics across Lattice.
1
0
3
Achieve sub-second temporal alignment between IMU signals and video pose sequences by focusing on skeletal motion rather than raw pixels, enabling more accurate cross-modal understanding.
