Depthwise Convolution Is All You Need for Learning Multiple Visual Domains.

Yunhui Guo, Yandong Li, Liqiang Wang, Tajana Rosing

Browse the full AAAI paper archive.