Research map / Detection and segmentation / Weakly supervised detection and segmentation
Consistency-based Semi-supervised Learning for Object detection
Read the paperCheck 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 |
|---|---|---|---|
| add CSD consistency on unlabeled VOC12 | SSD512 · VOC07 test (VOC07 L + VOC12 U) | mAP: 73.3 -> 75.8 | Helped |
| add CSD consistency on unlabeled VOC12 | R-FCN · VOC07 test (VOC07 L + VOC12 U) | mAP: 73.9 -> 74.7 | Helped |
| add Background Elimination (ablation) | SSD512 CSD · VOC07 test (VOC07 L + VOC12 U) | mAP: 74.8 -> 75.8 | Helped |
| add Background Elimination (ablation) | SSD300 CSD · VOC07 test (VOC07 L + VOC12 U) | mAP: 72.0 -> 72.3 | Helped |
| use stricter BE mask (mk ⊗ mk′) (ablation) | SSD300 CSD · VOC07 test (VOC07 L + VOC12 U) | mAP: 72.3 -> 71.7 | Hurt |
| use L2-norm instead of JS divergence for classification consistency (ablation) | SSD300 CSD · VOC07 test (VOC07 L + VOC12 U) | mAP: 71.6 -> 70.0 | Hurt |
| add localization consistency to classification consistency (no BE) (ablation) | SSD300 CSD · VOC07 test (VOC07 L + VOC12 U) | mAP: 71.6 -> 72.0 | Helped |
| add unlabeled COCO (all 80 classes) to VOC12 (ablation) | SSD512 CSD · VOC07 test | mAP: 75.8 -> 75.1 | Hurt |
| add unlabeled COCO (20 VOC classes) to VOC12 (ablation) | R-FCN CSD · VOC07 test | mAP: 74.7 -> 75.1 | Helped |
| apply consistency loss on labeled data only (ablation) | detectors · VOC07 | mAP | Hurt |
Closest papers by meaning
- Multi-clue Consistency Learning to Bridge Gaps Between General and Oriented Object in Semi-supervised Detection (AAAI 2025 · similarity 0.91)
- Semi-DETR: Semi-Supervised Object Detection With Detection Transformers (CVPR 2023 · similarity 0.90)
- Consistent-Teacher: Towards Reducing Inconsistent Pseudo-Targets in Semi-Supervised Object Detection (CVPR 2023 · similarity 0.90)
- Unbiased Teacher v2: Semi-Supervised Object Detection for Anchor-Free and Anchor-Based Detectors (CVPR 2022 · similarity 0.89)
- Sparse Semi-DETR: Sparse Learnable Queries for Semi-Supervised Object Detection (CVPR 2024 · similarity 0.86)
- 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.