Research map / Detection and segmentation / CNN features
Side-Aware Boundary Localization for More Precise Object Detection
Jiaqi Wang, Wenwei Zhang, Yuhang Cao, Kai Chen, Jiangmiao Pang, Tao Gong, Jianping Shi, Chen Change Loy, Dahua Lin
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 SABL to RetinaNet | RetinaNet R-101 · COCO test-dev | AP: 1.7% | Improved |
| add SABL to Faster R-CNN | Faster R-CNN R-101 · COCO | AP: 3.0% | Improved |
| add SABL to Cascade R-CNN | Cascade R-CNN R-101 · COCO test-dev | AP: 0.9% | Improved |
| use boundary regression targets instead of bounding box regression (ablation) | Faster R-CNN R-50 · COCO val | AP: 36.4 -> 37.3 | Improved |
| add side-aware feature extraction (SAFE) (ablation) | Faster R-CNN R-50 with boundary regression · COCO val | AP: 37.3 -> 38.5 | Improved |
| use heavier 4conv1fc head instead of SAFE (ablation) | Faster R-CNN with boundary regression · COCO val | AP: 37.3 -> 37.4 | No clear effect |
| replace bounding box regression with BLB (ablation) | Faster R-CNN R-50 · COCO val | AP: 36.4 -> 38.3 | Improved |
| add BLB to SAFE (ablation) | Faster R-CNN R-50 · COCO val | AP: 38.5 -> 39.0 | Improved |
| add bucketing-guided rescoring (BGR) (ablation) | SABL · COCO val | AP: 39.0 -> 39.7 | Improved |
| bucketing without fine regression (ablation) | Faster R-CNN R-50 · COCO val | AP: 36.4 -> 32.8 | Hurt |
| localize object center with bucketing instead of boundaries (CLB vs BLB) (ablation) | Faster R-CNN R-50 · COCO val | AP: 38.3 -> 36.9 | Hurt |
| attention mask instead of max pooling for aggregating RoI features (ablation) | SAFE · COCO val | AP: 39.4 -> 39.7 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.
- Cascade R-CNN — Cascade R-CNN: Delving Into High Quality Object Detection (CVPR 2018) · better on 1 of 2 shared results
- Faster R-CNN — Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks (TPAMI 2016) · better on 2 of 2 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
- Bounding Box Regression With Uncertainty for Accurate Object Detection (CVPR 2019 · similarity 0.91)
- Dive Deeper Into Box for Object Detection (ECCV 2020 · similarity 0.88)
- Robust Object Detection with Inaccurate Bounding Boxes (ECCV 2022 · similarity 0.87)
- LocNet: Improving Localization Accuracy for Object Detection (CVPR 2016 · similarity 0.87)
- Theoretically Achieving Continuous Representation of Oriented Bounding Boxes (CVPR 2024 · similarity 0.85)
- Recognize Any Regions (NeurIPS 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.
