SGV3D: Toward Scenario Generalization for Vision-Based Roadside 3D Object Detection
- 2 June 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 26 (8) , 11782-11793
Abstract
Roadside perception can significantly enhance the safety of autonomous vehicles by extending their perceptual capabilities beyond the visual range and addressing occluded regions. However, current state-of-the-art vision-based roadside detection methods exhibit high accuracy on labeled scenes but perform poorly on new scenes. This limitation arises because roadside cameras remain stationary after installation and can only gather data from a single scene, leading the algorithm to overfit these roadside backgrounds and camera positions. To tackle this issue, we propose an innovative Scenario Generalization Framework for Vision-based Roadside 3D Object Detection, called SGV3D. Specifically, we utilize a Background-suppressed Module (BSM) to reduce background overfitting in vision-centric pipelines by diminishing background features during the 2D to bird’s-eye-view projection. Furthermore, by introducing the Semi-supervised Data Generation Pipeline (SSDG) that employs unlabeled images from new scenes, we generate diverse foreground instances with varying camera poses, mitigating the risk of overfitting to specific camera positions. Experiments conducted on two large-scale roadside benchmarks demonstrate that SGV3D, with only a minimal increase in latency, effectively improves the scenario generalization capabilities of vision-based roadside 3D object detectors. The code is available here (https://github.com/yanglei18/SGV3D).Keywords
Funding Information
- National Natural Science Foundation of China (52221005, 52072215, 62273198, U1964203)
- Beijing Natural Science Foundation (L241017, L243025)
- National Key Research and Development Program of China (2022YFB2503003)
This publication has 45 references indexed in Scilit:
- GraphBEV: Towards Robust BEV Feature Alignment for Multi-modal 3D Object Detection2024
- Lite-FPN for keypoint-based monocular 3D object detectionKnowledge-Based Systems, 2023
- BEVFormer: Learning Bird’s-Eye-View Representation from Multi-camera Images via Spatiotemporal Transformers2022
- Rope3D: The Roadside Perception Dataset for Autonomous Driving and Monocular 3D Object Detection Task2022
- DAIR-V2X: A Large-Scale Dataset for Vehicle-Infrastructure Cooperative 3D Object Detection2022
- Delving into Localization Errors for Monocular 3D Object Detection2021
- Categorical Depth Distribution Network for Monocular 3D Object Detection2021
- Kinematic 3D Object Detection in Monocular Video2020
- Deep Residual Learning for Image Recognition2016
- Are we ready for autonomous driving? The KITTI vision benchmark suite2012