SR-BEVS：稀疏恢复 BEV 模型汤提升相机 3D 目标检测鲁棒性

View PDF HTML (experimental)

Abstract:Camera-based bird's-eye view (BEV) perception methods such as Fast-BEV have demonstrated promising results with fast inference speed and competitive accuracy. However, their robustness to real-world corruptions remains insufficiently explored, posing risks for autonomous driving deployment where environmental variations are inevitable. In this paper, we propose Sparse Recovery Bird's-Eye View Model Soup (SR-BEVS), a sparse recovery learning approach that simultaneously improves clean accuracy and corruption robustness without requiring additional data or architectural modifications. Our method identifies pruning candidates based on layer-wise weight magnitudes, applies random pruning to create diverse sparse models, and retrains each model with a randomly sampled learning rate. By averaging these diverse recovered models into a single dense model, we achieve enhanced robustness against various corruptions while introducing zero inference overhead. We further provide a mode connectivity analysis demonstrating that our pruning-based diversification preserves the connected low-loss basin necessary for stable weight averaging. Experiments on the nuScenes dataset demonstrate up to 33.8% improvement in corruption robustness and a 1.8% gain in clean accuracy over the Fast-BEV baseline, with consistent improvements across sensor noise, weather distortions, and illumination changes. Our simple yet effective approach may be applicable to other BEV models.

Comments:	Accepted to IROS 2026. 8 pages, 6 figures
Subjects:	Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Cite as:	arXiv:2610.34976 [cs.CV]
	(or arXiv:2610.34976v1 [cs.CV] for this version)
	https://doi.org/10.48550/arXiv.2610.34976 arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Jinwoo Chung [view email]
[v1] Mon, 26 Oct 2026 07:49:42 UTC (16,594 KB)