kevin-ssy/CLIP_as_RNN

Official Implementation for CVPR 2024 paper: CLIP as RNN: Segment Countless Visual Concepts without Training Endeavor