Research map / Detection and segmentation / CNN features
CentripetalNet: Pursuing High-Quality Keypoint Pairs for Object Detection
Zhiwei Dong, Guoxuan Li, Yue Liao, Fei Wang, Pengju Ren, Chen Qian
Check 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 |
|---|---|---|---|
| Replace associative embedding with centripetal shift matching (ablation) | CornerNet · MS-COCO test-dev, CornerNet matching ablation | AP: 37.5 -> 40.7 | Improved |
| Add center prediction and CenterNet matching (ablation) | CornerNet · MS-COCO test-dev, CornerNet matching ablation | AP: 37.5 -> 39.9 | Improved |
| Replace associative embedding with center regression (ablation) | CornerNet · MS-COCO test-dev, CornerNet matching ablation | AP: 37.5 -> 40.1 | Improved |
| Add standard deformable convolution (ablation) | CentripetalNet without feature adaptation or mask head · MS-COCO test-dev, feature adaptation ablation | AP: 40.7 -> 40.8 | Improved |
| Use RoI convolution for feature adaptation (ablation) | CentripetalNet without feature adaptation or mask head · MS-COCO test-dev, feature adaptation ablation | AP: 40.7 -> 41.1 | Improved |
| Add cross-star deformable convolution guided by corner-to-center shifts (ablation) | CentripetalNet without feature adaptation or mask head · MS-COCO test-dev, feature adaptation ablation | AP: 40.7 -> 41.5 | Improved |
| Add the mask prediction module (ablation) | CentripetalNet · MS-COCO test-dev, 110 epochs | AP: 41.5 -> 41.8 | Improved |
| Add the mask prediction module (ablation) | CentripetalNet · MS-COCO test-dev, 210 epochs | AP: 41.7 -> 42.1 | Improved |
| Add a mask prediction module (ablation) | CornerNet · MS-COCO test-dev, 110 epochs | AP: 37.3 -> 37.3 | No clear effect |
| Use multi-scale rather than single-scale testing | CentripetalNet · MS-COCO test-dev2017 object detection | AP: 46.1 -> 48.0 | Improved |
Later papers that compared with this method
3 later accepted papers report CentripetalNet as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- Gradient Corner Pooling for Keypoint-Based Object Detection (AAAI 2023 · better on 1 of 1 shared result)
- Cornerformer: Purifying Instances for Corner-Based Detectors (ECCV 2022 · better on 2 of 2 shared results)
- HumanLiker: A Human-like Object Detector to Model the Manual Labeling Process (NeurIPS 2022 · better on 1 of 1 shared result)
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.
- CenterNet511-104 — CenterNet: Keypoint Triplets for Object Detection (ICCV 2019) · better on 1 of 1 shared result
- PANet — PANet: Few-Shot Image Semantic Segmentation With Prototype Alignment (ICCV 2019) · better on 0 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
- CenterNet: Keypoint Triplets for Object Detection (ICCV 2019 · similarity 0.90)
- CornerNet: Detecting Objects as Paired Keypoints (ECCV 2018 · similarity 0.87)
- CenterNet++ for Object Detection (TPAMI 2023 · similarity 0.87)
- PolarNet: Learning to Optimize Polar Keypoints for Keypoint Based Object Detection (ICLR 2021 · similarity 0.86)
- Sim2Real Object-Centric Keypoint Detection and Description (AAAI 2022 · similarity 0.86)
- Towards Complementary Keypoint Detection via Mixture of Detectors (NeurIPS 2026 · 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.
