PP-YOLO: An Effective and Efficient Implementation of Object Detector

TL;DR

PP-YOLO is an improved version of YOLOv3 that combines multiple optimization tricks to enhance accuracy while maintaining real-time speed, making it suitable for practical applications.

Contribution

The paper introduces PP-YOLO, a practical object detector based on YOLOv3 that improves accuracy without increasing model complexity, using various existing tricks.

Findings

Achieves 45.2% mAP, surpassing state-of-the-art detectors.

Maintains high inference speed at 72.9 FPS.

Effectively balances accuracy and efficiency for real-world use.

Abstract

Object detection is one of the most important areas in computer vision, which plays a key role in various practical scenarios. Due to limitation of hardware, it is often necessary to sacrifice accuracy to ensure the infer speed of the detector in practice. Therefore, the balance between effectiveness and efficiency of object detector must be considered. The goal of this paper is to implement an object detector with relatively balanced effectiveness and efficiency that can be directly applied in actual application scenarios, rather than propose a novel detection model. Considering that YOLOv3 has been widely used in practice, we develop a new object detector based on YOLOv3. We mainly try to combine various existing tricks that almost not increase the number of model parameters and FLOPs, to achieve the goal of improving the accuracy of detector as much as possible while ensuring that the speed is almost unchanged. Since all experiments in this paper are conducted based on PaddlePaddle, we call it PP-YOLO. By combining multiple tricks, PP-YOLO can achieve a better balance between effectiveness (45.2% mAP) and efficiency (72.9 FPS), surpassing the existing state-of-the-art detectors such as EfficientDet and YOLOv4.Source code is at https://github.com/PaddlePaddle/PaddleDetection.