Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Reliability-Aware Prediction via Uncertainty Learning for Person Image Retrieval
Zhaopeng Dou, Zhongdao Wang, Weihua Chen, Yali Li, Shengjin Wang
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 data uncertainty learning (ablation) | UAL (ResNet50) · MSMT17 | mAP (%): 52.0 -> 53.1 | Improved |
| add model uncertainty learning (ablation) | UAL (ResNet50) · MSMT17 | mAP (%): 52.0 -> 52.3 | Improved |
| add model uncertainty on top of data uncertainty (ablation) | UAL (ResNet50) · MSMT17 | mAP (%): 53.1 -> 56.5 | Improved |
| add data uncertainty learning (ablation) | UAL (HRNet-W32) · MSMT17 | mAP (%): 59.6 -> 62.2 | Improved |
| add both uncertainties (ablation) | UAL (HRNet-W32) · MSMT17 | mAP (%): 59.6 -> 65.3 | Improved |
| add data uncertainty learning (ablation) | BOT(S50) · MSMT17 | mAP (%): 59.4 -> 59.5 | No clear effect |
| add both uncertainties (ablation) | BOT(R50) · MSMT17 | mAP (%): 49.9 -> 53.6 | Improved |
| add reliability score weighting of queries (ablation) | UAL multi-query · Market-1501 multi-query, Gaussian noise | mAP (%): 53.1 -> 71.4 | Improved |
| add reliability score weighting of queries (ablation) | UAL multi-query · MSMT17 multi-query, motion blur | mAP (%): 47.1 -> 54.9 | Improved |
| add reliability score weighting of queries (ablation) | UAL multi-query · Occluded-REID multi-query | mAP (%): 72.2 -> 73.5 | Improved |
| increase Bayesian samples T from 10 to 20 (ablation) | UAL ResNet50 · CUHK03 labeled | mAP (%): 75.6 -> 75.6 | No clear effect |
| reduce T from 10 to 5 (ablation) | UAL ResNet50 · Market-1501 | mAP (%): 87.0 -> 86.9 | No clear effect |
Closest papers by meaning
- Adaptive Uncertainty-Based Learning for Text-Based Person Retrieval (AAAI 2024 · similarity 0.89)
- Risk Controlled Image Retrieval (AAAI 2025 · similarity 0.87)
- Composed Image Retrieval with Text Feedback via Multi-grained Uncertainty Regularization (ICLR 2024 · similarity 0.85)
- SURE: SUrvey REcipes for building reliable and robust deep networks (CVPR 2024 · similarity 0.84)
- From Risk to Uncertainty: Generating Predictive Uncertainty Measures via Bayesian Estimation (ICLR 2025 · similarity 0.83)
- UncertainSAM: Fast and Efficient Uncertainty Quantification of the Segment Anything Model (ICML 2025 · similarity 0.83)
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.
