Research map / Vision-language and multimodal / Cross-modal learning
Geometry-Aware Noisy Correspondence Mitigation for Cross-Modal Text-Based Person Retrieval
Xinpan Yuan, Shaomin Xie, Liujie Hua, Chengyuan Zhang, Guihu Zhao, Lin Yuanbo Wu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add GSCA and SRAM to RDE | RDE · CUHK-PEDES (0% noise) | R@1: 75.94 -> 77.36 | Improved |
| add GSCA and SRAM to IRRA | IRRA · CUHK-PEDES (50% noise) | R@1: 62.41 -> 70.46 | Improved |
| add GSCA and SRAM to RDE | RDE · CUHK-PEDES (50% noise) | R@1: 71.25 -> 74.25 | Improved |
| add GSCA only to RDE (ablation) | RDE · CUHK-PEDES (50% noise) | R@1: 71.25 -> 73.46 | Improved |
| add SRAM only to RDE (ablation) | RDE · CUHK-PEDES (50% noise) | R@1: 71.25 -> 73.62 | Improved |
| combine GSCA with SRAM vs either alone (ablation) | RDE · CUHK-PEDES (50% noise) | R@1: 73.62 -> 74.25 | Improved |
| add GSCA only to RDE (clean) (ablation) | RDE · CUHK-PEDES (0% noise) | R@1: 75.94 -> 76.63 | Improved |
| add SRAM only to RDE (clean) (ablation) | RDE · CUHK-PEDES (0% noise) | R@1: 75.94 -> 76.47 | Improved |
| add GSCA and SRAM to IRRA | IRRA · ICFG-PEDES (50% noise) | mAP: 29.05 -> 25.47 | Hurt |
| add GSCA and SRAM to IRRA | IRRA · ICFG-PEDES (50% noise) | R@1: 52.53 -> 62.80 | Improved |
| add GSCA and SRAM to IRRA | IRRA · RSTPReid (50% noise) | R@1: 56.65 -> 59.72 | Improved |
| set cross-modal temperature 0.03, intra weight 0.4, SRAM temperature 0.07, threshold 0.5 (ablation) | GSCA and SRAM · CUHK-PEDES (50% noise) | R@1 | Improved |
Closest papers by meaning
- Adaptive Uncertainty-Based Learning for Text-Based Person Retrieval (AAAI 2024 · similarity 0.89)
- Mitigating Noisy Correspondence by Geometrical Structure Consistency Learning (CVPR 2024 · similarity 0.89)
- Noisy Correspondence Rectification via Asymmetric Similarity Learning (AAAI 2025 · similarity 0.88)
- Robust Noisy Correspondence Learning with Equivariant Similarity Consistency (CVPR 2024 · similarity 0.88)
- Causality-Inspired Invariant Representation Learning for Text-Based Person Retrieval (AAAI 2024 · similarity 0.87)
- Unifying Multi-Modal Uncertainty Modeling and Semantic Alignment for Text-to-Image Person Re-identification (AAAI 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
