论文标题
缩放Yolov4：缩放跨阶段部分网络
Scaled-YOLOv4: Scaling Cross Stage Partial Network
论文作者
论文摘要
我们表明，Yolov4对象检测神经网络基于CSP方法，在上下缩放，并且适用于小型和大型网络，同时保持最佳速度和准确性。我们提出了一种网络缩放方法，该方法不仅修改了深度，宽度，分辨率，还可以修改网络的结构。 Yolov4-Large模型可实现最先进的结果：MS可可数据集的55.5％AP（73.4％AP50）在Tesla V100上的速度约为16 fps，而随着测试时间的增加，Yolov4-large-Large实现了56.0％AP（73.3 AP50）。据我们所知，这是任何已发表工作中可可数据集上最高准确性的。 Yolov4微型模型在RTX 2080TI上以443 fps的速度实现22.0％的AP（42.0％AP50），而使用张力，批量尺寸= 4和FP16准则，Yolov4-Tiny Accon actiny Acthie actie and Acnie a achie fls 1774 fps。
We show that the YOLOv4 object detection neural network based on the CSP approach, scales both up and down and is applicable to small and large networks while maintaining optimal speed and accuracy. We propose a network scaling approach that modifies not only the depth, width, resolution, but also structure of the network. YOLOv4-large model achieves state-of-the-art results: 55.5% AP (73.4% AP50) for the MS COCO dataset at a speed of ~16 FPS on Tesla V100, while with the test time augmentation, YOLOv4-large achieves 56.0% AP (73.3 AP50). To the best of our knowledge, this is currently the highest accuracy on the COCO dataset among any published work. The YOLOv4-tiny model achieves 22.0% AP (42.0% AP50) at a speed of 443 FPS on RTX 2080Ti, while by using TensorRT, batch size = 4 and FP16-precision the YOLOv4-tiny achieves 1774 FPS.
