Towards Latency-Aware 3D Streaming Perception for Autonomous Driving

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Peng, Jiaqi, Wang, Tai, Pang, Jiangmiao, Shen, Yuan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916709177229312
author Peng, Jiaqi
Wang, Tai
Pang, Jiangmiao
Shen, Yuan
author_facet Peng, Jiaqi
Wang, Tai
Pang, Jiangmiao
Shen, Yuan
contents Although existing 3D perception algorithms have demonstrated significant improvements in performance, their deployment on edge devices continues to encounter critical challenges due to substantial runtime latency. We propose a new benchmark tailored for online evaluation by considering runtime latency. Based on the benchmark, we build a Latency-Aware 3D Streaming Perception (LASP) framework that addresses the latency issue through two primary components: 1) latency-aware history integration, which extends query propagation into a continuous process, ensuring the integration of historical feature regardless of varying latency; 2) latency-aware predictive detection, a module that compensates the detection results with the predicted trajectory and the posterior accessed latency. By incorporating the latency-aware mechanism, our method shows generalization across various latency levels, achieving an online performance that closely aligns with 80\% of its offline evaluation on the Jetson AGX Orin without any acceleration techniques.
format Preprint
id arxiv_https___arxiv_org_abs_2504_19115
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Towards Latency-Aware 3D Streaming Perception for Autonomous Driving
Peng, Jiaqi
Wang, Tai
Pang, Jiangmiao
Shen, Yuan
Computer Vision and Pattern Recognition
Robotics
Although existing 3D perception algorithms have demonstrated significant improvements in performance, their deployment on edge devices continues to encounter critical challenges due to substantial runtime latency. We propose a new benchmark tailored for online evaluation by considering runtime latency. Based on the benchmark, we build a Latency-Aware 3D Streaming Perception (LASP) framework that addresses the latency issue through two primary components: 1) latency-aware history integration, which extends query propagation into a continuous process, ensuring the integration of historical feature regardless of varying latency; 2) latency-aware predictive detection, a module that compensates the detection results with the predicted trajectory and the posterior accessed latency. By incorporating the latency-aware mechanism, our method shows generalization across various latency levels, achieving an online performance that closely aligns with 80\% of its offline evaluation on the Jetson AGX Orin without any acceleration techniques.
title Towards Latency-Aware 3D Streaming Perception for Autonomous Driving
topic Computer Vision and Pattern Recognition
Robotics
url https://arxiv.org/abs/2504.19115