YOLOv4: Optimal Speed and Accuracy of Object Detection

Abstract

There are a huge number of features which are said to improve Convolutional Neural Network (CNN) accuracy. Practical testing of combinations of such features on large datasets, and theoretical justification of the result, is required. Some features operate on certain models exclusively and for certain problems exclusively, or only for small-scale datasets; while some features, such as batch-normalization and residual-connections, are applicable to the majority of models, tasks, and datasets. We assume that such universal features include Weighted-Residual-Connections (WRC), Cross-Stage-Partial-connections (CSP), Cross mini-Batch Normalization (CmBN), Self-adversarial-training (SAT) and Mish-activation. We use new features: WRC, CSP, CmBN, SAT, Mish activation, Mosaic data augmentation, CmBN, DropBlock regularization, and CIoU loss, and combine some of them to achieve state-of-the-art results: 43.5% AP (65.7% AP50) for the MS COCO dataset at a realtime speed of ~65 FPS on Tesla V100. Source code is at https://github.com/AlexeyAB/darknet

References in corpus (8)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Improved Regularization of Convolutional Neural Networks with Cutout
• Searching for Activation Functions
• Random Erasing Data Augmentation
• Learning Spatial Fusion for Single-Shot Object Detection
• RetinaMask: Learning to predict masks improves state-of-the-art single-shot detection for free
• Label Refinement Network for Coarse-to-Fine Semantic Segmentation
• RDSNet: A New Deep Architecture for Reciprocal Object Detection and Instance Segmentation

Cited by in corpus (16)

• PP-YOLO: An Effective and Efficient Implementation of Object Detector
• A Vision-based Social Distancing and Critical Density Detection System for COVID-19
• A Comprehensive Survey of Machine Learning Applied to Radar Signal Processing
• Autonomous Driving with Deep Learning: A Survey of State-of-Art Technologies
• Dual Semantic Fusion Network for Video Object Detection
• Corner Proposal Network for Anchor-free, Two-stage Object Detection
• AReLU: Attention-based Rectified Linear Unit
• 1st Place Solution of LVIS Challenge 2020: A Good Box is not a Guarantee of a Good Mask
• AinnoSeg: Panoramic Segmentation with High Perfomance
• Sequence-guided protein structure determination using graph convolutional and recurrent networks
• Vec2Instance: Parameterization for Deep Instance Segmentation
• A Study on Evaluation Standard for Automatic Crack Detection Regard the Random Fractal
• 2nd Place Solution to ECCV 2020 VIPriors Object Detection Challenge
• A novel approach to remove foreign objects from chest X-ray images
• Actor-Action Video Classification CSC 249/449 Spring 2020 Challenge Report
• Object Detection-Based Variable Quantization Processing