YOLOv4: Optimal Speed and Accuracy of Object Detection

Alexey Bochkovskiy, Chien-Yao Wang, Hong-Yuan Mark Liao
4/23/2020
cs.CVeess.IV

Official code: AlexeyAB/darknet (22,145 stars, last commit 2025-12-15).

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

View on arXivView PDF

Code Implementations(1)

AlexeyAB/darknetOfficial100%

YOLOv4 / Scaled-YOLOv4 / YOLO - Neural Networks for Object Detection (Windows and Linux version of Darknet )

22,1457,841C, C#Dec 2, 201610 months agoNOASSERTION
computer-visiondeep-learningdeep-learning-tutorialdeep-neural-networksdnn+7 more

Similar Papers

GMA-SAWGAN-GP: A Novel Data Generative Framework to Enhance IDS Detection Performance

81% match2026

Attack Assessment and Augmented Identity Recognition for Human Skeleton Data

81% match2026

Multi-Granular Discretization for Interpretable Generalization in Precise Cyberattack Identification

81% match2025

Adaptive Meta-Learning for Robust Deepfake Detection: A Multi-Agent Framework to Data Drift and Model Generalization

80% match1 repo12024

Revisiting Video Quality Assessment from the Perspective of Generalization

80% match1 repo22024

ANROT-HELANet: Adverserially and Naturally Robust Attention-Based Aggregation Network via The Hellinger Distance for Few-Shot Classification

80% match1 repo2025

Cite this paper

@article{bochkovskiy2020yolov,
  title  = {YOLOv4: Optimal Speed and Accuracy of Object Detection},
  author = {Alexey Bochkovskiy and Chien-Yao Wang and Hong-Yuan Mark Liao},
  year   = {2020},
  eprint = {2004.10934},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/2004.10934}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2004.10934.svg)](https://codeofpaper.com/paper/2004.10934)

Discussion