CLIP as RNN: Segment Countless Visual Concepts without Training Endeavor.

Shuyang Sun, Runjia Li, Philip Torr, Xiuye Gu, Siyang Li

Browse the full CVPR paper archive.