PP-YOLOE: An evolved version of YOLO

TL;DR

PP-YOLOE is an advanced object detection model that improves accuracy and speed over previous YOLO versions, optimized for industrial deployment with various model sizes and high-performance inference capabilities.

Contribution

The paper introduces PP-YOLOE, a new state-of-the-art object detector with optimized architecture and training strategies, achieving higher accuracy and faster inference for industrial applications.

Findings

PP-YOLOE-l achieves 51.4 mAP on COCO test-dev.

PP-YOLOE-l runs at 78.1 FPS on Tesla V100.

Inference speed reaches 149.2 FPS with TensorRT and FP16.

Abstract

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.