Real-time Streaming Perception System for Autonomous Driving

Nowadays, plenty of deep learning technologies are being applied to all aspects of autonomous driving with promising results. Among them, object detection is the key to improve the ability of an autonomous agent to perceive its environment so that it can (re)act. However, previous vision-based object detectors cannot achieve satisfactory performance under real-time driving scenarios. To remedy this, we present the real-time streaming perception system in this paper, which is also the 2nd Place solution of Streaming Perception Challenge (Workshop on Autonomous Driving at CVPR 2021) for the detection-only track. Unlike traditional object detection challenges, streaming perception task requires achieving a balance of accuracy and latency rather than focusing on the absolute performance, which is crucial for real-time autonomous driving. Based on YOLOv5, we adopted data augmentation, Bag-of-Freebies, and Transformer to improve streaming object detection performance with negligible extra inference cost. On the Argoverse-HD test set, our method achieved 33.2 streaming AP (34.6 streaming AP verified by the organizer) under the required hardware. Its performance significantly surpasses the fixed baseline of 13.6 (host team), demonstrating the potentiality of application.