Research map / Detection and segmentation / CNN features
Reconcile Prediction Consistency for Balanced Object Detection
Keyang Wang, Lei Zhang
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 Harmonic loss (ablation) | Faster R-CNN · MS COCO val | AP: 37.3 -> 38.5 | Improved |
| add standard IoU loss to Harmonic loss (ablation) | Faster R-CNN · MS COCO val | AP: 38.5 -> 38.7 | Improved |
| add HIoU loss to Harmonic loss (ablation) | Faster R-CNN · MS COCO val | AP: 38.5 -> 39.2 | Improved |
| add Task-Contrastive loss to Harmonic loss (ablation) | Faster R-CNN · MS COCO val | AP: 38.3 -> 38.5 | Improved |
| raise HIoU gamma from 0.0 to 0.8 (ablation) | Faster R-CNN · MS COCO val | AP: 38.7 -> 39.2 | Improved |
| raise HIoU gamma from 0.8 to 1.0 (ablation) | Faster R-CNN · MS COCO val | AP: 39.2 -> 39.0 | Hurt |
| raise HIoU alpha from 1.5 to 2.5 (ablation) | Faster R-CNN · MS COCO val | AP: 39.2 -> 38.9 | Hurt |
| raise HIoU alpha from 0.5 to 1.5 (ablation) | Faster R-CNN · MS COCO val | AP: 38.9 -> 39.2 | Improved |
| add HarmonicDet loss | Faster R-CNN ResNet-50-FPN · PASCAL VOC 2007 test | AP: 51.7 -> 54.2 | Improved |
| add HarmonicDet loss | SSD300 · PASCAL VOC 2007 test | AP: 51.0 -> 54.5 | Improved |
| add HarmonicDet loss | RefineDet512 · PASCAL VOC 2007 test | AP: 56.3 -> 60.4 | Improved |
| add HarmonicDet loss | RefineDet512 · PASCAL VOC 2007 test | AP90: 19.6 -> 29.7 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Faster R-CNN — Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (TPAMI 2016) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Correlation Loss: Enforcing Correlation between Classification and Localization (AAAI 2023 · similarity 0.88)
- Towards Accurate One-Stage Object Detection With AP-Loss (CVPR 2019 · similarity 0.86)
- Acquisition of Localization Confidence for Accurate Object Detection (ECCV 2018 · similarity 0.86)
- The Equalization Losses: Gradient-Driven Training for Long-tailed Object Recognition (TPAMI 2023 · similarity 0.86)
- AP-Loss for Accurate One-Stage Object Detection (TPAMI 2020 · similarity 0.86)
- Improving Single Domain-Generalized Object Detection: A Focus on Diversification and Alignment (CVPR 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.
