Research map / Detection and segmentation / CNN features
Context Refinement for Object Detection
Zhe Chen, Shaoli Huang, Dacheng Tao
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 context refinement to the RPN (C-RPN) (ablation) | FRCNN† · VOC 07 test | AP@0.5: 0.796 -> 0.804 | Improved |
| add context refinement to the RPN (C-RPN) (ablation) | FRCNN† · VOC 07 test | AP@0.8: 0.442 -> 0.469 | Improved |
| add context refinement to the final refinement stage (ablation) | FRCNN† with C-RPN · VOC 07 test | AP@0.5: 0.804 -> 0.822 | Improved |
| add context refinement to the final refinement stage (ablation) | FRCNN† with C-RPN · VOC 07 test | AP@0.7: 0.650 -> 0.685 | Improved |
| add context refinement to Faster RCNN | FRCNN† VGG · VOC 07 test | mAP: 75.3 -> 77.6 | Improved |
| add context refinement to Faster RCNN | FRCNN† Res101 · VOC 07 test | mAP: 79.6 -> 82.2 | Improved |
| add context refinement to Faster RCNN | FRCNN† Res101 · MS COCO test-dev | AP: 37.5 -> 39.0 | Improved |
| add context refinement to Deformable R-FCN | DRFCN Res101 · MS COCO test-dev | AP: 37.1 -> 39.1 | Improved |
| add context refinement to Mask RCNN | Mask RCNN Res101 · MS COCO test-dev | AP: 40.2 -> 42.0 | Improved |
| add context refinement to Mask RCNN | Mask RCNN Res101 · MS COCO test-dev | mAP@0.7: 43.9 -> 46.4 | Improved |
| add context refinement to Faster RCNN | FRCNN† Res101 · MS COCO test-dev | mAP(large): 51.1 -> 53.0 | Improved |
| add context refinement to Faster RCNN (Res101) | FRCNN† Res101 · MS COCO test-dev | mAP(small): 18.8 -> 19.4 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- HyperNet — HyperNet: Towards Accurate Region Proposal Generation and Joint Object Detection (CVPR 2016) · better on 1 of 1 shared result
- R-FCN — R-FCN: Object Detection via Region-based Fully Convolutional Networks (NeurIPS 2016) · better on 1 of 1 shared result
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
- ContextShift: A Controlled Benchmark for Context Dependence in Object Detection (NeurIPS 2026 · similarity 0.90)
- Exploring the Bounds of the Utility of Context for Object Detection (CVPR 2019 · similarity 0.89)
- Enriched Feature Guided Refinement Network for Object Detection (ICCV 2019 · similarity 0.89)
- Seeing without Looking: Contextual Rescoring of Object Detections for AP Maximization (CVPR 2020 · similarity 0.88)
- Transformer-Based Context Condensation for Boosting Feature Pyramids in Object Detection (IJCV 2023 · similarity 0.88)
- Toward Scale-Invariance and Position-Sensitive Region Proposal Networks (ECCV 2018 · 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.
