Learning Spatio-Temporal Structure from RGB-D Videos for Human Activity Detection and Anticipation.

Hema Swetha Koppula, Ashutosh Saxena

Browse the full ICML paper archive.