In this report, we present some experienced improvements to YOLO series, forming a new high-performance detector -- YOLOX. We switch the YOLO detector to an anchor-free manner and conduct other advanced detection techniques, i.e., a decoupled head and the leading label assignment strategy SimOTA to achieve state-of-the-art results across a large scale range of models: For YOLO-Nano with only 0.91M parameters and 1.08G FLOPs, we get 25.3% AP on COCO, surpassing NanoDet by 1.8% AP; for YOLOv3, one of the most widely used detectors in industry, we boost it to 47.3% AP on COCO, outperforming the current best practice by 3.0% AP; for YOLOX-L with roughly the same amount of parameters as YOLOv4-CSP, YOLOv5-L, we achieve 50.0% AP on COCO at a speed of 68.9 FPS on Tesla V100, exceeding YOLOv5-L by 1.8% AP. Further, we won the 1st Place on Streaming Perception Challenge (Workshop on Autonomous Driving at CVPR 2021) using a single YOLOX-L model. We hope this report can provide useful experience for developers and researchers in practical scenes, and we also provide deploy versions with ONNX, TensorRT, NCNN, and Openvino supported. Source code is at https://github.com/Megvii-BaseDetection/YOLOX.
翻译:在本报告中,我们介绍了对YOLO系列的一些改进,形成了一个新的高性能探测器 -- -- YOLOX。我们将YOLO探测器转换为无锚制式,并开展了其他先进的探测技术,即分解头和领先标签分配战略SimOTA,以在大范围的模型中取得最先进的成果:对于YOLO-Nano,仅有0.91M参数和1.08G FLOPs,我们在COCO公司上获得了25.3%的AP,比NanoDet高出1.8%;对于YOLOV3,这是工业中最常用的探测器之一,我们将其提升到47.3%的AP,比目前的最佳做法高3.0% AP;对于YOLOX-L,其参数大致与YOLOv4-CSP、YOLOV5-L, 我们实现了50.OFPS, 以68.9的速度在Tesla V100的NEOLVVV-L,超过YOLOV-L, AP1.8%的部署速度超过YOL-APOVOL-DROVO, 我们还在AVOL AL VERVERO 上赢得了SOVOL-SO 20VOL-SOVERVOL 的S-SOVERVERVERVOR 20VOR 版本。