Research map / Detection and segmentation / Weakly supervised detection and segmentation
Calibrated Teacher for Sparsely Annotated Object Detection
Haohan Wang, Liang Liu, Boshen Zhang, Jiangning Zhang, Wuhao Zhang, Zhenye Gan, Yabiao Wang, Chengjie Wang, Haoqian Wang
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 |
|---|---|---|---|
| Use Calibrated Teacher and Focal IoU together | RetinaNet on COCO sparse splits · COCO Split-1, Split-2, and Split-3 | mAP | Improved |
| Add Calibrated Teacher (ablation) | RetinaNet · COCO Easy, Hard, and Extreme | mAP: 37.1 -> 37.7; 35.3 -> 35.5; 26.7 -> 27.4 | Improved |
| Add Focal IoU (ablation) | RetinaNet · COCO Easy, Hard, and Extreme | mAP: 37.1 -> 37.7; 35.3 -> 35.6; 26.7 -> 27.8 | Improved |
| Combine Calibrated Teacher with Focal IoU (ablation) | RetinaNet · COCO Easy, Hard, and Extreme | mAP: 37.1 -> 38.1; 35.3 -> 36.4; 26.7 -> 28.7 | Improved |
| Calibrate teacher confidence rather than use a fixed score threshold | RetinaNet, FCOS, and Faster R-CNN · COCO Split-1 50% | mAP: 38.5 vs 38.5; 38.7 -> 39.0; 38.6 -> 39.3 | Mixed |
| Vary calibrated score threshold across 0.55, 0.65, 0.7, 0.75, and 0.85 (ablation) | Calibrated Teacher · COCO Split-1 50% | mAP: 37.6; 38.3; 38.5; 38.1; 38.0 | No clear effect |
| Set Focal IoU w0 to 0 rather than 0.5 (ablation) | Focal IoU · COCO Split-1 50% | mAP: 39.3 -> 33.9 | Hurt |
| Vary Focal IoU k from 1.0 to 1.5 or 2.0 at w0=0.5 (ablation) | Focal IoU · COCO Split-1 50% | mAP: 39.2; 39.3; 39.3 | No clear effect |
| Increase Focal IoU w0 from 0.5 to 1 at k=1.5 (ablation) | Focal IoU · COCO Split-1 50% | mAP: 39.3 -> 39.3 | No clear effect |
| Reduce calibration queue size from 8000 to 4000 images of predictions (ablation) | Calibrated Teacher · COCO Hard | mAP: 33.0 -> 33.6 | Improved |
| Increase calibration queue size from 8000 to 16000 images of predictions (ablation) | Calibrated Teacher · COCO Hard | mAP: 33.0 -> 33.0 | No clear effect |
| Vary calibrator training interval T among 250, 500, and 1000 (ablation) | Calibrated Teacher · COCO Hard | mAP: 33.0; 33.0; 33.2 | No clear effect |
Closest papers by meaning
- Co-Student: Collaborating Strong and Weak Students for Sparsely Annotated Object Detection (ECCV 2024 · similarity 0.92)
- SparseDet: Improving Sparsely Annotated Object Detection with Pseudo-positive Mining (ICCV 2023 · similarity 0.90)
- Label, Verify, Correct: A Simple Few Shot Object Detection Method (CVPR 2022 · similarity 0.90)
- S²Teacher: Step-by-step Teacher for Sparsely Annotated Oriented Object Detection (AAAI 2026 · similarity 0.89)
- Many-shot from Low-shot: Learning to Annotate using Mixed Supervision for Object Detection (ECCV 2020 · similarity 0.89)
- Object Detection Benchmarks are Incomplete: The Role of Label Errors and Annotation Uncertainty (NeurIPS 2026 · similarity 0.87)
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.
