CentripetalNet: Pursuing High-quality Keypoint Pairs for Object Detection

Abstract

Keypoint-based detectors have achieved pretty-well performance. However, incorrect keypoint matching is still widespread and greatly affects the performance of the detector. In this paper, we propose CentripetalNet which uses centripetal shift to pair corner keypoints from the same instance. CentripetalNet predicts the position and the centripetal shift of the corner points and matches corners whose shifted results are aligned. Combining position information, our approach matches corner points more accurately than the conventional embedding approaches do. Corner pooling extracts information inside the bounding boxes onto the border. To make this information more aware at the corners, we design a cross-star deformable convolution network to conduct feature adaption. Furthermore, we explore instance segmentation on anchor-free detectors by equipping our CentripetalNet with a mask prediction module. On MS-COCO test-dev, our CentripetalNet not only outperforms all existing anchor-free detectors with an AP of 48.0% but also achieves comparable performance to the state-of-the-art instance segmentation approaches with a 40.2% MaskAP. Code will be available at https://github.com/KiveeDong/CentripetalNet.

Accepted by CVPR2020

References in corpus (14)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• YOLOv3: An Incremental Improvement
• FCOS: Fully Convolutional One-Stage Object Detection
• DenseBox: Unifying Landmark Localization with End to End Object Detection
• Path Aggregation Network for Instance Segmentation
• Hybrid Task Cascade for Instance Segmentation
• CenterNet: Keypoint Triplets for Object Detection
• RetinaMask: Learning to predict masks improves state-of-the-art single-shot detection for free
• Bottom-up Object Detection by Grouping Extreme and Center Points
• Scale-Aware Trident Networks for Object Detection
• Revisiting Feature Alignment for One-stage Object Detection
• PolarMask: Single Shot Instance Segmentation with Polar Representation
• Region Proposal by Guided Anchoring
• Acquisition of Localization Confidence for Accurate Object Detection

Cited by in corpus (7)

• Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detection
• Probabilistic two-stage detection
• RelationNet++: Bridging Visual Representations for Object Detection via Transformer Decoder
• Autonomous Driving with Deep Learning: A Survey of State-of-Art Technologies
• PPDM: Parallel Point Detection and Matching for Real-time Human-Object Interaction Detection
• Generalized Focal Loss V2: Learning Reliable Localization Quality Estimation for Dense Object Detection
• SiamCorners: Siamese Corner Networks for Visual Tracking