Deep Evidential Learning with Noisy Correspondence for Cross-Modal Retrieval

Benchmark Model Rank Results
cross-modal-retrieval-with-noisy-1DECL-SGRAF#10R-Sum: 364.3Image-to-text R@1: 39.0Image-to-text R@5: 66.1
cross-modal-retrieval-with-noisy-2DECL-SGRAF#10R-Sum: 494.7Image-to-text R@1: 77.5Image-to-text R@5: 93.8
cross-modal-retrieval-with-noisy-3DECL-SGARF#11R-Sum: 518.2Image-to-text R@1: 77.5Image-to-text R@5: 95.9
text-based-person-retrieval-with-noisyDECL#2Rank-1: 70.29Rank 10: 91.93Rank-5: 87.04mAP: 62.84mINP: 46.54
text-based-person-retrieval-with-noisy-1DECL#2Rank 1: 61.95Rank-10: 83.88Rank-5: 78.36mAP: 36.08mINP: 6.25
text-based-person-retrieval-with-noisy-2DECL#2Rank 1: 61.75Rank 10: 86.90Rank 5: 80.70mAP: 47.70mINP: 26.07