首页|期刊导航|Computers, Materials & Continua|Real-Time 3D Scene Perception in Dynamic Urban Environments via Street Detection Gaussians

Real-Time 3D Scene Perception in Dynamic Urban Environments via Street Detection GaussiansOA

中文摘要

As a cornerstone for applications such as autonomous driving,3D urban perception is a burgeoning field of study.Enhancing the performance and robustness of these perception systems is crucial for ensuring the safety of next-generation autonomous vehicles.In this work,we introduce a novel neural scene representation called Street Detection Gaussians(SDGs),which redefines urban 3D perception through an integrated architecture unifying reconstruction and detection.At its core lies the dynamic Gaussian representation,where time-conditioned parameterization enables simultaneous modeling of static environments and dynamic objects through physically constrained Gaussian evolution.The framework’s radar-enhanced perception module learns cross-modal correlations between sparse radardata anddense visual features,resulting ina22%reduction inocclusionerrors compared tovisiononly systems.A breakthrough differentiable rendering pipeline back-propagates semantic detection losses throughout the entire 3D reconstruction process,enabling the optimization of both geometric and semantic fidelity.Evaluated on the Waymo Open Dataset and the KITTI Dataset,the system achieves real-time performance(135 Frames Per Second(FPS)),photorealistic quality(Peak Signal-to-Noise Ratio(PSNR)34.9 dB),and state-of-the-art detection accuracy(78.1%Mean Average Precision(mAP)),demonstrating a 3.8×end-to-end improvement over existing hybrid approaches while enabling seamless integration with autonomous driving stacks.

Yu Du;Runwei Guan;Ho-Pun Lam;Jeremy Smith;Yutao Yue;KaLok Man;Yan Li

School of Advanced Technology,Xi’an Jiaotong-Liverpool University,Suzhou,215123,ChinaThrust of Artificial Intelligence,The Hong Kong University of Science and Technology(Guangzhou),Guangzhou,511400,ChinaSchool of Advanced Technology,Xi’an Jiaotong-Liverpool University,Suzhou,215123,ChinaDepartment of Electrical Engineering and Electronics,University of Liverpool,Liverpool,L697ZX,UKThe Hong Kong University of Science and Technology(Guangzhou),Guangzhou 511400,China Institute of Deep Perception Technology,JITRI,Wuxi,214000,ChinaSchool of Advanced Technology,Xi’an Jiaotong-Liverpool University,Suzhou,215123,ChinaDepartment of Electrical and Computer Engineering,Inha University,Incheon,402751,Republic of Korea

交通工程

Radar-vision fusiondifferentiable renderingautonomous driving perception3D reconstructionocclusion robustness

《Computers, Materials & Continua》 2026 (4)

P.1384-1402,19

10.32604/cmc.2025.072544

评论