Research map / Label-efficient and robust learning / Domain adaptation and generalization
Automatic Adaptation of Object Detectors to New Domains Using Self-Training
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Retrain on high-confidence pseudo-labels from target detections | WIDER baseline · CS6-Test | AP: 15.66 -> 17.29 | Helped |
| Add tracker-derived hard positives to detection pseudo-labels | WIDER baseline · CS6-Test | AP: 17.29 -> 17.31 | Helped |
| Use tracker-only pseudo-labels instead of detector-derived labels (ablation) | WIDER baseline · CS6-Test | AP: 15.66 -> 11.73 | Hurt |
| Replace hard labels with interpolated soft labels (ablation) | CS6-HP · CS6-Test | AP: 17.31 -> 19.89 (λ=0.3); 19.56 (λ=0.5); 20.80 (λ=0.7) | Helped |
| Use source-to-target score histogram remapping for soft labels (ablation) | WIDER baseline · CS6-Test | AP: 17.31 -> 19.12 | Helped |
| Emphasize tracker-only examples while constraining detector-sample scores (ablation) | WIDER baseline · CS6-Test | AP: 17.31 -> 20.65 | Helped |
| Add ROI-level discriminator and consistency term to image-level domain adaptation (ablation) | CS6-DA-im · CS6-Test | AP: 21.02 -> 22.18 | Helped |
| Retrain on high-confidence pseudo-labels from target detections | BDD source baseline · BDD-Target-Test | AP: 15.21 -> 26.16 | Helped |
| Add tracker-derived hard positives to detection pseudo-labels | BDD source baseline · BDD-Target-Test | AP: 26.16 -> 27.11 | Helped |
| Replace hard labels with interpolated soft labels (ablation) | BDD-HP · BDD-Target-Test | AP: 27.11 -> 28.59 (λ=0.3); 28.38 (λ=0.5); 28.47 (λ=0.7) | Helped |
| Use source-to-target score histogram remapping for soft labels (ablation) | BDD source baseline · BDD-Target-Test | AP: 27.11 -> 28.02 | Helped |
| Emphasize tracker-only examples while constraining detector-sample scores (ablation) | BDD source baseline · BDD-Target-Test | AP: 27.11 -> 28.43 | Helped |
Closest papers by meaning
- Deep Face Detector Adaptation Without Negative Transfer or Catastrophic Forgetting (CVPR 2018 · similarity 0.88)
- Object Detection With Self-Supervised Scene Adaptation (CVPR 2023 · similarity 0.88)
- LEOD: Label-Efficient Object Detection for Event Cameras (CVPR 2024 · similarity 0.86)
- Test-Time Adaptive Object Detection with Foundation Model (NeurIPS 2025 · similarity 0.85)
- Large Self-Supervised Models Bridge the Gap in Domain Adaptive Object Detection (CVPR 2025 · similarity 0.85)
- DetKDS: Knowledge Distillation Search for Object Detectors (ICML 2024 · similarity 0.85)
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.