China
Hangzhou Dianzi University
Unsupervised video object segmentation, long temporal memory, short temporal attention, efficient projection