Research map / Label-efficient and robust learning / Recognition and long-tail
BIT: Matching-based Bi-directional Interaction Transformation Network for Visible-Infrared Person Re-Identification
Haoxuan Xu, Guanglin Niu
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 Bi-directional Cross Interaction (BCI) module to the PMT baseline (ablation) | PMT baseline · SYSU-MM01 All-Search Single-Shot | Rank-1 / mAP (%): 69.23 -> 75.24 Rank-1; 66.02 -> 73.35 mAP | Improved |
| add Aggregation Contrastive Loss LAC after BCI (ablation) | BIT · SYSU-MM01 All-Search Single-Shot | Rank-1 / mAP (%): 75.24 -> 76.42 Rank-1; 73.35 -> 74.54 mAP | Improved |
| add Query Aware Scoring module (ablation) | BIT · SYSU-MM01 All-Search Single-Shot | Rank-1 / mAP (%): 75.24 -> 79.53 Rank-1; 73.35 -> 79.02 mAP | Improved |
| add LAC on top of BCI and QA Scoring (full model) (ablation) | BIT · SYSU-MM01 All-Search Single-Shot | Rank-1 (%): 79.53 -> 80.53 | Improved |
| use standard cross-attention in place of BCI (ablation) | PMT baseline · SYSU-MM01 All-Search Single-Shot | Rank-1 / mAP (%): 69.23 -> 68.68 Rank-1; 66.02 -> 64.67 mAP | Hurt |
| use bi-directional BCI instead of standard cross-attention (ablation) | PMT baseline · SYSU-MM01 All-Search Single-Shot | Rank-1 (%): 68.68 -> 75.24 | Improved |
| set Top-k to 3 (ablation) | QA Scoring · SYSU-MM01 All-Search Single-Shot | Rank-1 (%) | Improved |
| set penalty coefficient alpha to 0.2 and loss weight lambda to 0.6 (ablation) | BIT · SYSU-MM01 All-Search Single-Shot | Rank-1 (%) | Improved |
| add re-ranking to BIT | BIT · SYSU-MM01 All-Search Single-Shot | Rank-1 (%): 80.53 -> 84.42 | Improved |
| use coarse candidate filtering at inference with K=50 | BIT · SYSU-MM01 inference latency | latency per query: 10ms (PMT) vs 14ms (BIT) | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- WRIM-Net — WRIM-Net: Wide-Ranging Information Mining Network for Visible-Infrared Person Re-Identification (ECCV 2024) · better on 2 of 2 shared results
- PartMix — PartMix: Regularization Strategy To Learn Part Discovery for Visible-Infrared Person Re-Identification (CVPR 2023) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Learning by Aligning: Visible-Infrared Person Re-Identification Using Cross-Modal Correspondences (ICCV 2021 · similarity 0.92)
- Modality Unifying Network for Visible-Infrared Person Re-Identification (ICCV 2023 · similarity 0.91)
- Empowering Visible-Infrared Person Re-Identification with Large Foundation Models (NeurIPS 2024 · similarity 0.91)
- TokenMatcher: Diverse Tokens Matching for Unsupervised Visible-Infrared Person Re-Identification (AAAI 2025 · similarity 0.90)
- High-Order Structure Based Middle-Feature Learning for Visible-Infrared Person Re-identification (AAAI 2024 · similarity 0.90)
- Shallow-Deep Collaborative Learning for Unsupervised Visible-Infrared Person Re-Identification (CVPR 2024 · similarity 0.90)
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.
