Figure-ground segmentation improves handled object recognition in egocentric video.

Xiaofeng Ren, Chunhui Gu

Browse the full CVPR paper archive.