arXiv cs.AI· Jinwoo Chung, Sangho An, Hyunsung Park, Jangho Kim·· 15 小时前AI 评分22
SR-BEVS：面向鲁棒纯视觉 3D 目标检测的稀疏恢复 BEV 模型汤
Sparse Recovery Bird's-Eye View Model Soup for Robust Camera-Based 3D Object Detection
AI 导读
研究者提出 SR-BEVS，一种稀疏恢复学习式的模型汤方法，可在不增加数据和改动架构的前提下同时提升 Fast-BEV 的干净精度与损坏鲁棒性。该方法按层权重幅值筛选剪枝候选、随机剪枝生成多样稀疏模型并各自以随机采样学习率重训，再平均回单个稠密模型，推理零额外开销。在 nuScenes 上，损坏鲁棒性最高提升 33.8%，干净精度提升 1.8%，对传感器噪声、天气失真和光照变化均有一致改善。
正文 · 原文
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)
来源：arXiv cs.AI · arxiv.org