CenterNet: Keypoint Triplets for Object Detection

Abstract

In object detection, keypoint-based approaches often suffer a large number of incorrect object bounding boxes, arguably due to the lack of an additional look into the cropped regions. This paper presents an efficient solution which explores the visual patterns within each cropped region with minimal costs. We build our framework upon a representative one-stage keypoint-based detector named CornerNet. Our approach, named CenterNet, detects each object as a triplet, rather than a pair, of keypoints, which improves both precision and recall. Accordingly, we design two customized modules named cascade corner pooling and center pooling, which play the roles of enriching information collected by both top-left and bottom-right corners and providing more recognizable information at the central regions, respectively. On the MS-COCO dataset, CenterNet achieves an AP of 47.0%, which outperforms all existing one-stage detectors by at least 4.9%. Meanwhile, with a faster inference speed, CenterNet demonstrates quite comparable performance to the top-ranked two-stage detectors. Code is available at https://github.com/Duankaiwen/CenterNet.

10 pages (including 2 pages of References), 7 figures, 5 tables

References in corpus (5)

• R-FCN: Object Detection via Region-based Fully Convolutional Networks
• DSSD : Deconvolutional Single Shot Detector
• Beyond Skip Connections: Top-Down Modulation for Object Detection
• Scale-Aware Trident Networks for Object Detection
• ME R-CNN: Multi-Expert R-CNN for Object Detection

Cited by in corpus (39)

• A Survey of Deep Learning-based Object Detection
• Recent advances and clinical applications of deep learning in medical image analysis
• Deep High-Resolution Representation Learning for Visual Recognition
• Deep learning for detection and segmentation of artefact and disease instances in gastrointestinal endoscopy
• Review of data analysis in vision inspection of power lines with an in-depth discussion of deep learning technology
• RepPoints V2: Verification Meets Regression for Object Detection
• Revisiting Feature Alignment for One-stage Object Detection
• Decoupled Classification Refinement: Hard False Positive Suppression for Object Detection
• Detection of 3D Bounding Boxes of Vehicles Using Perspective Transformation for Accurate Speed Measurement
• Physical Adversarial Attack on Vehicle Detector in the Carla Simulator
• PVN3D: A Deep Point-wise 3D Keypoints Voting Network for 6DoF Pose Estimation
• Learning Human-Object Interaction Detection using Interaction Points
• CentripetalNet: Pursuing High-quality Keypoint Pairs for Object Detection
• Generalized Focal Loss V2: Learning Reliable Localization Quality Estimation for Dense Object Detection
• A Real-Time Cross-modality Correlation Filtering Method for Referring Expression Comprehension
• Progressive DARTS: Bridging the Optimization Gap for NAS in the Wild
• Dense RepPoints: Representing Visual Objects with Dense Point Sets
• Real-time 3D Deep Multi-Camera Tracking
• Associate-3Ddet: Perceptual-to-Conceptual Association for 3D Point Cloud Object Detection
• Objects in Semantic Topology
• Occluded Prohibited Items Detection: an X-ray Security Inspection Benchmark and De-occlusion Attention Module
• A Survey on Deep Domain Adaptation and Tiny Object Detection Challenges, Techniques and Datasets
• AFP-Net: Realtime Anchor-Free Polyp Detection in Colonoscopy
• TubeTK: Adopting Tubes to Track Multi-Object in a One-Step Training Model
• PAFNet: An Efficient Anchor-Free Object Detector Guidance
• Extreme Consistency: Overcoming Annotation Scarcity and Domain Shifts
• Learning to Cluster Faces via Confidence and Connectivity Estimation
• MatrixNets: A New Scale and Aspect Ratio Aware Architecture for Object Detection
• FlatteNet: A Simple Versatile Framework for Dense Pixelwise Prediction
• Egoshots, an ego-vision life-logging dataset and semantic fidelity metric to evaluate diversity in image captioning models
• Dually Supervised Feature Pyramid for Object Detection and Segmentation
• More Reliable AI Solution: Breast Ultrasound Diagnosis Using Multi-AI Combination
• Semi-Anchored Detector for One-Stage Object Detection
• Visible Feature Guidance for Crowd Pedestrian Detection
• An Experience Report on Machine Learning Reproducibility: Guidance for Practitioners and TensorFlow Model Garden Contributors
• Seq-SG2SL: Inferring Semantic Layout from Scene Graph Through Sequence to Sequence Learning
• DSIC: Dynamic Sample-Individualized Connector for Multi-Scale Object Detection
• Lesion-Inspired Denoising Network: Connecting Medical Image Denoising and Lesion Detection
• Survey on Deep Learning-based Kuzushiji Recognition