CornerNet-Lite: Efficient Keypoint Based Object Detection

Abstract

Keypoint-based methods are a relatively new paradigm in object detection, eliminating the need for anchor boxes and offering a simplified detection framework. Keypoint-based CornerNet achieves state of the art accuracy among single-stage detectors. However, this accuracy comes at high processing cost. In this work, we tackle the problem of efficient keypoint-based object detection and introduce CornerNet-Lite. CornerNet-Lite is a combination of two efficient variants of CornerNet: CornerNet-Saccade, which uses an attention mechanism to eliminate the need for exhaustively processing all pixels of the image, and CornerNet-Squeeze, which introduces a new compact backbone architecture. Together these two variants address the two critical use cases in efficient object detection: improving efficiency without sacrificing accuracy, and improving accuracy at real-time efficiency. CornerNet-Saccade is suitable for offline processing, improving the efficiency of CornerNet by 6.0x and the AP by 1.0% on COCO. CornerNet-Squeeze is suitable for real-time detection, improving both the efficiency and accuracy of the popular real-time detector YOLOv3 (34.4% AP at 30ms for CornerNet-Squeeze compared to 33.0% AP at 39ms for YOLOv3 on COCO). Together these contributions for the first time reveal the potential of keypoint-based detection to be useful for applications requiring processing efficiency.

arXiv commentAccepted to BMVC 2020

References in corpus (15)

• Adam: A Method for Stochastic Optimization
• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
• YOLOv3: An Incremental Improvement
• R-FCN: Object Detection via Region-based Fully Convolutional Networks
• DSSD : Deconvolutional Single Shot Detector
• Mixed Precision Training
• Pruning Filters for Efficient ConvNets
• Associative Embedding: End-to-End Learning for Joint Detection and Grouping
• Beyond Skip Connections: Top-Down Modulation for Object Detection
• Light-Head R-CNN: In Defense of Two-Stage Object Detector
• Deformable ConvNets v2: More Deformable, Better Results
• Point Linking Network for Object Detection

Cited by in corpus (27)

• YOLOv4: Optimal Speed and Accuracy of Object Detection
• A Survey of Deep Learning-based Object Detection
• CSPNet: A New Backbone that can Enhance Learning Capability of CNN
• A Survey of Convolutional Neural Networks: Analysis, Applications, and Prospects
• Scaled-YOLOv4: Scaling Cross Stage Partial Network
• Fewer is More: Efficient Object Detection in Large Aerial Images
• Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selection
• Continuous Human Action Recognition for Human-Machine Interaction: A Review
• Rethinking Classification and Localization for Object Detection
• RelationNet++: Bridging Visual Representations for Object Detection via Transformer Decoder
• MiniVLM: A Smaller and Faster Vision-Language Model
• YOLObile: Real-Time Object Detection on Mobile Devices via Compression-Compilation Co-Design
• Corner Proposal Network for Anchor-free, Two-stage Object Detection
• RetinaTrack: Online Single Stage Joint Detection and Tracking
• Occluded Prohibited Items Detection: an X-ray Security Inspection Benchmark and De-occlusion Attention Module
• 6-PACK: Category-level 6D Pose Tracker with Anchor-Based Keypoints
• Long-term Human Motion Prediction with Scene Context
• SiamCorners: Siamese Corner Networks for Visual Tracking
• Testing Deep Learning Models for Image Analysis Using Object-Relevant Metamorphic Relations
• MatrixNets: A New Scale and Aspect Ratio Aware Architecture for Object Detection
• Adaptive Class Suppression Loss for Long-Tail Object Detection
• SNIDER: Single Noisy Image Denoising and Rectification for Improving License Plate Recognition
• CoDeNet: Efficient Deployment of Input-Adaptive Object Detection on Embedded FPGAs
• Towards Understanding the Effectiveness of Attention Mechanism
• Localize to Classify and Classify to Localize: Mutual Guidance in Object Detection
• Pixel-Semantic Revise of Position Learning A One-Stage Object Detector with A Shared Encoder-Decoder
• AGSFCOS: Based on attention mechanism and Scale-Equalizing pyramid network of object detection