Learning to Generate Diverse Pedestrian Movements from Web Videos with Noisy Labels.

Zhizheng Liu, Joe Lin, Wayne Wu, Bolei Zhou

Browse the full ICLR paper archive.