论文标题
PP-Yoloe：Yolo的演变版本
PP-YOLOE: An evolved version of YOLO
论文作者
论文摘要
在本报告中，我们提出了PP-Yoloe，这是一种具有高性能和友好部署的工业最先进的对象探测器。我们使用不含锚固范式，更强大的骨干和颈部配备了CSPREPRESSTAGE，ET-HEAD和动态标签分配算法TAL的基础，更强大的骨干和颈部优化。我们为不同的实践场景提供S/M/L/X模型。结果，与先前的pp-yoloe-l在Coco Test-DEV上达到51.4映射，而Tesla V100上的78.1 fps则达到了78.1 fps，与先前的The-Art Industrip pp-yolov2相比，与先前的状态pp-yolov2相比，（+1.9 ap， +13.35％， +13.35％速度）和（+1.3 ap， +1.3 ap， +1.24.96％速度）相比。此外，PP-YOLOE推理速度以张力和FP16精确率达到149.2 fps。我们还进行了广泛的实验，以验证设计的有效性。源代码和预培训模型可在https://github.com/paddlepaddle/paddledetection上找到。
In this report, we present PP-YOLOE, an industrial state-of-the-art object detector with high performance and friendly deployment. We optimize on the basis of the previous PP-YOLOv2, using anchor-free paradigm, more powerful backbone and neck equipped with CSPRepResStage, ET-head and dynamic label assignment algorithm TAL. We provide s/m/l/x models for different practice scenarios. As a result, PP-YOLOE-l achieves 51.4 mAP on COCO test-dev and 78.1 FPS on Tesla V100, yielding a remarkable improvement of (+1.9 AP, +13.35% speed up) and (+1.3 AP, +24.96% speed up), compared to the previous state-of-the-art industrial models PP-YOLOv2 and YOLOX respectively. Further, PP-YOLOE inference speed achieves 149.2 FPS with TensorRT and FP16-precision. We also conduct extensive experiments to verify the effectiveness of our designs. Source code and pre-trained models are available at https://github.com/PaddlePaddle/PaddleDetection.
