Dual-Prompt CLIP with Hybrid Visual Encoders for Occluded Person Re-Identification

Benchmark Model Rank Results
person-re-identification-on-occluded-dukemtmcDPL-ReID (CNN)–Rank-1: 70.6mAP: 57.2
person-re-identification-on-occluded-dukemtmcDPL-ReID (ViT)–Rank-1: 76.3mAP: 67.2