Research map / Label-efficient and robust learning / Semi-supervised learning
SJDL-Vehicle: Semi-supervised Joint Defogging Learning for Foggy Vehicle Re-identification
Wei-Ting Chen, I-Hsiang Chen, Chih-Yuan Yeh, Hao-Hsiang Yang, Jian-Jiun Ding, Sy-Yen Kuo
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 |
|---|---|---|---|
| Jointly train the defogging and ReID branches (ablation) | ResNet-50 vehicle ReID · FVRID_Real | mAP: 76.17 -> 79.47 | Improved |
| Add semi-supervised optimization on real-world foggy images (ablation) | JDL · FVRID_Real | mAP: 79.47 -> 82.70 | Improved |
| Add semi-supervised optimization on real-world foggy images (ablation) | JDL · FVRID_Syn | mAP: 83.04 -> 85.36 | Improved |
| Add the self-constraint loss LSC (ablation) | SJDL-Vehicle · FVRID_Real | mAP: 81.50 -> 82.70 | Improved |
| Add the self-constraint loss LSC (ablation) | SJDL-Vehicle · FVRID_Syn | mAP: 84.39 -> 85.36 | Improved |
| Use Conv_2 for the collective feature sharing module instead of Conv_3 (ablation) | SJDL-Vehicle · FVRID_Real | mAP: 81.94 -> 82.70 | Improved |
| Use Conv_2 for the collective feature sharing module instead of Conv_4 (ablation) | SJDL-Vehicle · FVRID_Real | mAP: 81.57 -> 82.70 | Improved |
| Use Conv_2 for the collective feature sharing module instead of Conv_5 (ablation) | SJDL-Vehicle · FVRID_Real | mAP: 79.39 -> 82.70 | Improved |
| Train the ReID model on foggy data rather than clear data | PVEN · FVRID_Real | mAP: 75.36 -> 81.92 | Improved |
| Apply two-stage defogging before ReID | PVEN · FVRID_Real | mAP: 75.36 -> 78.13 | Improved |
| Apply two-stage defogging before ReID | TransReID · FVRID_Real | mAP: 64.00 -> 65.30 | Improved |
| Apply two-stage defogging before ReID | VOC · FVRID_Real | mAP: 57.40 -> 49.20 | Hurt |
Closest papers by meaning
- RVSL: Robust Vehicle Similarity Learning in Real Hazy Scenes Based on Semi-Supervised Learning (ECCV 2022 · similarity 0.90)
- Disentangled Feature Learning Network and a Comprehensive Benchmark for Vehicle Re-Identification (TPAMI 2021 · similarity 0.88)
- Semantic Foggy Scene Understanding with Synthetic Data (IJCV 2018 · similarity 0.86)
- SynFog: A Photo-realistic Synthetic Fog Dataset based on End-to-end Imaging Simulation for Advancing Real-World Defogging in Autonomous Driving (CVPR 2024 · similarity 0.83)
- Day-Night Cross-domain Vehicle Re-identification (CVPR 2024 · similarity 0.83)
- Learning Generalized Segmentation for Foggy-Scenes by Bi-directional Wavelet Guidance (AAAI 2024 · similarity 0.82)
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.
