CLIP-Guided Cross-Modal Feature Fusion based Few-Shot Learning for Nighttime Pavement Defect Detection.

Chongchong Zan, Songlin Du, Takeshi Ikenaga

Browse the full MVA paper archive.