论文标题
YOLOV4：对象检测的最佳速度和准确性
YOLOv4: Optimal Speed and Accuracy of Object Detection
论文作者
论文摘要
据说有大量功能可以提高卷积神经网络（CNN）精度。需要对大型数据集中此类特征组合的实际测试，以及结果的理论理由。某些功能专门在某些模型上运行，仅在某些问题上或仅在小规模数据集上运行；虽然某些功能（例如批处理规范化和残差连接）适用于大多数模型，任务和数据集。我们假设这种通用特征包括加权 - 分离 - 连接（WRC），跨阶段 - 派对 - 连接（CSP），跨微型批准归一化（CMBN），自我逆向训练（SAT）和Mish-Activation。我们使用新功能：WRC，CSP，CMBN，SAT，MISH激活，Mosaic数据增强，CMBN，Dropblock正则化和CIOU损失，并将其中一些结合在一起以实现MS Coco DataSet的最新结果：43.5％AP（65.7％AP50），以实际的MS Coco DataSet的实时速度速度为〜65 fps v1 sella v100 sesla v100 sesla v100 sesla v1 00。源代码在https://github.com/alexeyab/darknet上
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
