Research map / Vision-language and multimodal / Cross-modal learning
Multilingual Text-to-Image Person Retrieval via Bidirectional Relation Reasoning and Aligning
Min Cao, Xinyu Zhou, Ding Jiang, Bo Du, Mang Ye, Min Zhang
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-lingual MLM and cross-lingual D-MIM (Bi-IRR) (ablation) | Bi-IRRA · CUHK-PEDES(M) English | R@1 (%): 76.97 -> 78.82 | Improved |
| add bi-lingual MLM (ablation) | Bi-IRRA · CUHK-PEDES(M) English | R@1 (%): 77.81 -> 78.82 | Improved |
| add cross-lingual D-MIM (ablation) | Bi-IRRA · CUHK-PEDES(M) English | R@1 (%): 77.49 -> 78.82 | Improved |
| replace SDM with Md-GA global alignment (ablation) | Bi-IRRA · CUHK-PEDES(M) English | R@1 (%): 66.89 -> 78.82 | Improved |
| add ID loss (ablation) | Bi-IRRA · CUHK-PEDES(M) English | R@1 (%): 78.82 -> 78.28 | No clear effect |
| use LDAT pipeline instead of direct LLM translation (ablation) | Bi-IRRA · CUHK-PEDES(M) English | R@1 (%): 78.09 -> 78.82 | Improved |
| use MLLM instead of LLM for direct translation (ablation) | Bi-IRRA · CUHK-PEDES(M) English | R@1 (%): 78.09 -> 78.61 | Improved |
| use MLLM instead of LLM in rewriting phase of LDAT (ablation) | LDAT · CUHK-PEDES(M) English | R@1 (%): 78.22 -> 78.82 | Improved |
| use blockwise masking instead of random masking in D-MIM (ablation) | Bi-IRRA · CUHK-PEDES(M) English | R@1 (%): 78.17 -> 78.82 | Improved |
| use blockwise masking instead of gridwise masking (ablation) | Bi-IRRA · CUHK-PEDES(M) English | R@1 (%): 77.89 -> 78.82 | Improved |
| use feature distillation with source-text teacher and target-text student, instead of swapped roles (ablation) | Bi-IRRA · CUHK-PEDES(M) English | R@1 (%): 78.59 -> 78.82 | Improved |
| mask the image paired with target text in A-ITM (ablation) | Bi-IRRA · CUHK-PEDES(M) English | R@1 (%): 77.92 -> 78.82 | Improved |
Closest papers by meaning
- Chat-based Person Retrieval via Dialogue-Refined Cross-Modal Alignment (CVPR 2025 · similarity 0.87)
- Adaptive Uncertainty-Based Learning for Text-Based Person Retrieval (AAAI 2024 · similarity 0.87)
- Dynamic Adapter with Semantics Disentangling for Cross-lingual Cross-modal Retrieval (AAAI 2025 · similarity 0.87)
- Beyond Global Similarity: Multi-Conditional Retrieval for Fine-Grained Cross-Modal Understanding (CVPR 2026 · similarity 0.86)
- Unifying Multi-Modal Uncertainty Modeling and Semantic Alignment for Text-to-Image Person Re-identification (AAAI 2024 · similarity 0.86)
- Automatic Synthetic Data and Fine-grained Adaptive Feature Alignment for Composed Person Retrieval (NeurIPS 2025 · similarity 0.86)
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.
