Research map / Detection and segmentation / Weakly supervised detection and segmentation
Dense Learning Based Semi-Supervised Object Detection
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 |
|---|---|---|---|
| Use DSL instead of supervised-only FCOS | FCOS · MS-COCO 10% labeled | mAP: 23.86 -> 36.22 | Helped |
| Use DSL instead of the strongest listed baseline | object detection · PASCAL-VOC07 test, unlabeled VOC12 | AP50:90: 48.7 -> 56.8 | Helped |
| Add Adaptive Filtering (ablation) | DSL on FCOS · MS-COCO 10% labeled | mAP: 23.7 -> 32.2 | Helped |
| Add MetaNet pseudo-label refinement (ablation) | DSL on FCOS · MS-COCO 10% labeled | mAP: 32.2 -> 32.5 | Helped |
| Add Aggregated Teacher (ablation) | DSL on FCOS · MS-COCO 10% labeled | mAP: 32.5 -> 34.5 | Helped |
| Add patch-shuffle augmentation (ablation) | DSL on FCOS · MS-COCO 10% labeled | mAP: 34.5 -> 34.9 | Helped |
| Add scale uncertainty-consistency loss (ablation) | DSL on FCOS · MS-COCO 10% labeled | mAP: 34.9 -> 36.2 | Helped |
| Use a single threshold instead of multi-level Adaptive Filtering (ablation) | FCOS pseudo-label assignment · MS-COCO 10% labeled | mAP: 36.2 vs. 27.1, 28.8, 30.7, 27.5 | Hurt |
| Use class-adaptive τ2 instead of fixed τ2=0.3 (ablation) | Adaptive Filtering · MS-COCO 10% labeled | mAP: 36.0 -> 36.2 | Helped |
| Add EMA parameter aggregation (ablation) | FCOS teacher · MS-COCO 10% labeled | mAP: 33.0 -> 34.1 | Helped |
| Add layer aggregation (ablation) | FCOS teacher · MS-COCO 10% labeled | mAP: 33.0 -> 35.0 | Helped |
| Combine EMA and layer aggregation (ablation) | FCOS teacher · MS-COCO 10% labeled | mAP: 35.0 -> 36.2 | Helped |
Closest papers by meaning
- DTG-SSOD: Dense Teacher Guidance for Semi-Supervised Object Detection (NeurIPS 2022 · similarity 0.92)
- SOOD: Towards Semi-Supervised Oriented Object Detection (CVPR 2023 · similarity 0.90)
- Sparse Semi-DETR: Sparse Learnable Queries for Semi-Supervised Object Detection (CVPR 2024 · similarity 0.88)
- Multi-clue Consistency Learning to Bridge Gaps Between General and Oriented Object in Semi-supervised Detection (AAAI 2025 · similarity 0.87)
- Semi-supervised Open-World Object Detection (AAAI 2024 · similarity 0.87)
- Relational Matching for Weakly Semi-Supervised Oriented Object Detection (CVPR 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.