Research map / Detection and segmentation / Open-set and few-shot detection
Black-Box Domain Adaptation for Object Detection with Retention-Driven Knowledge Compression
Yuwu Lu, Chunzhi Liu
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 Scene Compression on top of Memory Retention (ablation) | RDKC · Cityscapes -> Foggy-Cityscapes | mAP: 37.6 -> 40.7 | Improved |
| add Scene Compression on top of Memory Retention (ablation) | RDKC · Sim10K -> Cityscapes | AP of Car: 43.9 -> 49.1 | Improved |
| add Scene Compression on top of Memory Retention (ablation) | RDKC · KITTI -> Cityscapes | AP of Car: 47.1 -> 50.4 | Improved |
| add Scene Compression on top of Memory Retention (ablation) | RDKC · Pascal-VOC -> Watercolor | mAP: 50.1 -> 53.7 | Improved |
| add Memory Retention to LPLD adapted to BBDA (ablation) | LPLD · Cityscapes -> Foggy-Cityscapes | mAP: 32.6 -> 39.9 | Improved |
| add Memory Retention to IRG adapted to BBDA (ablation) | IRG · Cityscapes -> Foggy-Cityscapes | mAP: 33.1 -> 38.3 | Improved |
| add Memory Retention to LPLD adapted to BBDA (ablation) | LPLD · Pascal-VOC -> Watercolor | mAP: 50.7 -> 53.1 | Improved |
| add Memory Retention to LPLD adapted to BBDA (ablation) | LPLD · Sim10K -> Cityscapes | AP of Car: 38.9 -> 46.8 | Improved |
| set partition threshold lambda to 0 (all proposals low-reliability) (ablation) | RDKC · Cityscapes -> Foggy-Cityscapes | accuracy (mAP): 40.7 -> 25.1 | Hurt |
| set partition threshold lambda to 1.0 (all proposals high-reliability) (ablation) | RDKC · Cityscapes -> Foggy-Cityscapes | accuracy (mAP): 40.7 -> 36.4 | Hurt |
| RDKC vs previous BBDA method DINE | RDKC · Pascal-VOC -> Watercolor | mAP: +8.1 | Improved |
Closest papers by meaning
- Orthogonal Knowledge Refreshing for Domain-Incremental Object Detection (ECCV 2026 · similarity 0.88)
- SEEN-DA: SEmantic ENtropy guided Domain-aware Attention for Domain Adaptive Object Detection (CVPR 2025 · similarity 0.87)
- Diversify and Match: A Domain Adaptive Representation Learning Paradigm for Object Detection (CVPR 2019 · similarity 0.87)
- DA-Ada: Learning Domain-Aware Adapter for Domain Adaptive Object Detection (NeurIPS 2024 · similarity 0.86)
- Non-exemplar Domain Incremental Object Detection via Learning Domain Bias (AAAI 2024 · similarity 0.86)
- A Separation and Alignment Framework for Black-Box Domain Adaptation (AAAI 2024 · 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.
