3D Pose from Motion for Cross-View Action Recognition via Non-linear Circulant Temporal Encoding.

Ankur Gupta, Julieta Martinez, James J. Little, Robert J. Woodham

Browse the full CVPR paper archive.