YOLOv4: Optimal Speed and Accuracy of Object Detection
2020/04/23 by Alexey Bochkovskiy, Chien-Yao Wang, Bochkovskiy, Alexey +3 · 2 voices · 110 citations
Computer Science · Engineering · #Advanced Neural Network Applications #CCD and CMOS Imaging Sensors #Industrial Vision Systems and Defect Detection #cs.CV #eess.IV
paper · pdf · doi:10.48550/arxiv.2004.10934
openalex publication_date 2020/04/23 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/30
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
Citations
Cited by
- SynSur: An end-to-end generative pipeline for synthetic industrial surface defect generation and detection
- Attention from Above: A Multimodal Model for Drone-Based Object Localization
- Leveraging Dissimilarity Invariance as a Robust Anchor for Learning with Noisy Labels
- Toward Optimal Adenovirus Detection Using YOLO26
- IoUCert: Robustness Verification for Anchor-based Object Detectors
- Cognitive-YOLO: LLM-Driven Architecture Synthesis from First Principles of Data for Object Detection
- Training-Free Metrics for Synthetic Object Detection Data: A Proxy for Detector Performance
- Intentional Electromagnetic Interference Attacks on Facial Recognition
- Advancements in automated nuclei segmentation for histopathology using you only look once-driven approaches: A systematic review
- DeFloMat: Detection with Flow Matching for Stable and Efficient Generative Object Localization
- Evaluating an Adaptive Multispectral Turret System for Autonomous Tracking Across Variable Illumination Conditions
- Reading Legends on Ancient Coins: An Object Detection Approach for Character Recognition on a Novel Roman Republican Dataset
- The World is Your Canvas: Painting Promptable Events with Reference Images, Trajectories, and Text
- YOLO11-4K: An Efficient Architecture for Real-Time Small Object Detection in 4K Panoramic Images
- KLO-Net: A Dynamic K-NN Attention U-Net with CSP Encoder for Efficient Prostate Gland Segmentation from MRI
- VajraV1 -- The most accurate Real Time Object Detector of the YOLO family
- LiM-YOLO: Less is More with Pyramid Level Shift for Ship Detection in Optical Remote Sensing
- Transformer-Driven Multimodal Fusion for Explainable Suspiciousness Estimation in Visual Surveillance
- Towards Robust DeepFake Detection under Unstable Face Sequences: Adaptive Sparse Graph Embedding with Order-Free Representation and Explicit Laplacian Spectral Prior
- A Patient-Doctor-NLP-System to contest inequality for less privileged
- MT-Depth: Multi-task Instance feature analysis for the Depth Completion
- Real-time Cricket Sorting By Sex
- FOD-S2R: A FOD Dataset for Sim2Real Transfer Learning based Object Detection
- DAONet-YOLOv8: An Occlusion-Aware Dual-Attention Network for Tea Leaf Pest and Disease Detection
- Intelligent Image Search Algorithms Fusing Visual Large Models
- Multimodal Real-Time Anomaly Detection and Industrial Applications
- Dendritic Convolution for Noise Image Recognition
- T2I-Based Physical-World Appearance Attack against Traffic Sign Recognition Systems in Autonomous Driving
- MCAQ-YOLO: Morphological Complexity-Aware Quantization for Efficient Object Detection with Curriculum Learning
- YOLO-Drone: An Efficient Object Detection Approach Using the GhostHead Network for Drone Images
- Scale-Aware Relay and Scale-Adaptive Loss for Tiny Object Detection in Aerial Images
- SFFR: Spatial-Frequency Feature Reconstruction for Multispectral Aerial Object Detection
- A Visual Perception-Based Tunable Framework and Evaluation Benchmark for H.265/HEVC ROI Encryption
- Semantic-Guided Natural Language and Visual Fusion for Cross-Modal Interaction Based on Tiny Object Detection
- Object-Centric 3D Gaussian Splatting for Strawberry Plant Reconstruction and Phenotyping
- DARTS: A Drone-Based AI-Powered Real-Time Traffic Incident Detection System
- VISAT: Benchmarking Adversarial and Distribution Shift Robustness in Traffic Sign Recognition with Visual Attributes
- Emission-Forecasting-Based Spatial-Temporal Carbon Response: A Multi-Agent Attention-Enhanced Deep Learning Framework
- High-throughput Verticillium wilt detection in cotton: A comparative study of faster R-CNN and YOLOv11
- GaTector+: A Unified Head-free Framework for Gaze Object and Gaze Following Prediction
- Mask-Robust Face Verification for Online Learning via YOLOv5 and Residual Networks
- OOS-DSD: Improving Out-of-stock Detection in Retail Images using Auxiliary Tasks
- Rethinking Inference Placement for Deep Learning across Edge and Cloud Platforms: A Multi-Objective Optimization Perspective and Future Directions
- Radar-Camera Fused Multi-Object Tracking: Online Calibration and Common Feature
- Beyond Frequency: Scoring-Driven Debiasing for Object Detection via Blueprint-Prompted Image Synthesis
- ArmFormer: Lightweight Transformer Architecture for Real-Time Multi-Class Weapon Segmentation and Classification
- VAR-SLAM: Visual Adaptive and Robust SLAM for Dynamic Environments
- ReCon: Region-Controllable Data Augmentation with Rectification and Alignment for Object Detection
- A Modular Object Detection System for Humanoid Robots Using YOLO
- A Modular AIoT Framework for Low-Latency Real-Time Robotic Teleoperation in Smart Cities
- When Does Supervised Training Pay Off? The Hidden Economics of Object Detection in the Era of Vision-Language Models
- Source-Free Object Detection with Detection Transformer
- Layout-Independent License Plate Recognition via Integrated Vision and Language Models
- Uncertainty-Aware Post-Detection Framework for Enhanced Fire and Smoke Detection in Compact Deep Learning Models
- Automated Defect Detection for Mass-Produced Electronic Components Based on YOLO Object Detection Models
- MaizeStandCounting (MaSC): Automated and Accurate Maize Stand Counting from UAV Imagery Using Image Processing and Deep Learning
- Data Factory with Minimal Human Effort Using VLMs
- Ultralytics YOLO Evolution: An Overview of YOLO26, YOLO11, YOLOv8 and YOLOv5 Object Detectors for Computer Vision and Pattern Recognition
- Digital Twins for Intelligent Intersections: A Literature Review
- Semantic Visual Simultaneous Localization and Mapping: A Survey on State of the Art, Challenges, and Future Directions
- YOLO-Based Defect Detection for Metal Sheets
- YOLO26: Key Architectural Enhancements and Performance Benchmarking for Real-Time Object Detection
- HierLight-YOLO: A Hierarchical and Lightweight Object Detection Network for UAV Photography
- MS-YOLO: Infrared Object Detection for Edge Deployment via MobileNetV4 and SlideLoss
- Filtering with Confidence: When Data Augmentation Meets Conformal Prediction
- Neptune-X: Active X-to-Maritime Generation for Universal Maritime Object Detection
- Advancing Metallic Surface Defect Detection via Anomaly-Guided Pretraining on a Large Industrial Dataset
- Maize Seedling Detection Dataset (MSDD): A Curated High-Resolution RGB Dataset for Seedling Maize Detection and Benchmarking with YOLOv9, YOLO11, YOLOv12 and Faster-RCNN
- A Synthetic Dataset for Manometry Recognition in Robotic Applications
- A Novel Compression Framework for YOLOv8: Achieving Real-Time Aerial Object Detection on Edge Devices via Structured Pruning and Channel-Wise Distillation
- Attention-Enhanced Learning for Sensing-Assisted Long-Term Beam Tracking in mmWave Communications
- Knowledge Distillation for Sensing-Assisted Long-Term Beam Tracking in mmWave Communications
- FPI-Det: a face--phone Interaction Dataset for phone-use detection and understanding
- A New Hybrid Model of Generative Adversarial Network and You Only Look Once Algorithm for Automatic License-Plate Recognition
- An Analysis of Layer-Freezing Strategies for Enhanced Transfer Learning in YOLO Architectures
- TriLiteNet: Lightweight Model for Multi-Task Visual Perception
- Ensemble YOLO Framework for Multi-Domain Mitotic Figure Detection in Histopathology Images
- AReLU: Attention-based Rectified Linear Unit
- SAR-NAS: Lightweight SAR Object Detection with Neural Architecture Search
- Probabilistic two-stage detection
- YOLOX: Exceeding YOLO Series in 2021
- diveXplore 6.0: ITEC's Interactive Video Exploration System at VBS 2022
- Vec2Instance: Parameterization for Deep Instance Segmentation
- A Survey on Automated Driving System Testing: Landscapes and Trends
- Quantization Robustness to Input Degradations for Object Detection
- FlowDet: Overcoming Perspective and Scale Challenges in Real-Time End-to-End Traffic Detection
- Achieving Real-Time Object Detection on MobileDevices with Neural Pruning Search
- YOLObile: Real-Time Object Detection on Mobile Devices via Compression-Compilation Co-Design
- RICO: Two Realistic Benchmarks and an In-Depth Analysis for Incremental Learning in Object Detection
- A Cost-Effective Framework for Predicting Parking Availability Using Geospatial Data and Machine Learning
- TACR-YOLO: A Real-time Detection Framework for Abnormal Human Behaviors Enhanced with Coordinate and Task-Aware Representations
- Colon Polyps Detection from Colonoscopy Images Using Deep Learning
- Towards Powerful and Practical Patch Attacks for 2D Object Detection in Autonomous Driving
- The Role of Radiographic Knee Alignment in Total Knee Replacement Outcomes and Opportunities for Artificial Intelligence-Driven Assessment
- Improving Robustness of Learning-based Autonomous Steering Using Adversarial Images
- Non-anchor-based vehicle detection for traffic surveillance using bounding ellipses
- Real Time Child Abduction And Detection System
- Video-based Person Re-identification without Bells and Whistles
- Surface Water Cleaning Robots: A Structured Review of Technologies, Implementation Challenges, and Future Research Directions
- DoorDet: Semi-Automated Multi-Class Door Detection Dataset via Object Detection and Large Language Models
- LF-YOLO: A Lighter and Faster YOLO for Weld Defect Detection of X-ray Image
- Domain Adaptive YOLO for One-Stage Cross-Domain Detection
- Head Anchor Enhanced Detection and Association for Crowded Pedestrian Tracking
- Pulmonary embolism identification in computerized tomography pulmonary angiography scans with deep learning technologies in COVID-19 patients
- YOLOv8-Based Deep Learning Model for Automated Poultry Disease Detection and Health Monitoring paper
- Self-Supervised YOLO: Leveraging Contrastive Learning for Label-Efficient Object Detection
- YOLOv1 to YOLOv11: A Comprehensive Survey of Real-Time Object Detection Innovations and Challenges
- Revisiting Adversarial Patch Defenses on Object Detectors: Unified Evaluation, Large-Scale Dataset, and New Insights
- Small object detection [wikipedia]
- You Only Look Once [wikipedia]
Discussions
Related