激光雷达提示的时空多视图立体用于自动驾驶
LiDAR Prompted Spatio-Temporal Multi-View Stereo for Autonomous Driving
将稀疏但度量准确的激光雷达观测作为几何提示（硬先验锚定代价体、软特征引导）融入多视图立体，并用时空解码器融合相邻帧时序上下文，为自动驾驶场景下的度量深度估计提供可复用的融合范式。
原文依据 Sparse but metrically accurate LiDAR observations can serve as geometric prompts to anchor depth estimation in absolute scale
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论准确的度量深度对于自动驾驶感知与仿真至关重要，然而现有方法难以同时实现高度量精度、多视图与时序一致性以及跨域泛化。为应对这些挑战，本文提出 DriveMVS，一种新颖的多视图立体框架，通过两个关键洞见调和上述相互竞争的目标：(1) 稀疏但度量准确的激光雷达观测可作为几何提示，将深度估计锚定在绝对尺度上；(2) 对多种线索的深度融合对于消除歧义和增强鲁棒性至关重要，而时空解码器确保跨帧一致性。基于这些原则，DriveMVS 以两种方式嵌入激光雷达提示：作为锚定代价体的硬几何先验，以及作为由三线索组合器融合的软特征级引导。在时序一致性方面，DriveMVS 采用时空解码器，联合利用来自 MVS 代价体的几何线索和来自相邻帧的时序上下文。实验表明，DriveMVS 在多个基准上取得最优性能，在度量精度、时序稳定性和零样本跨域迁移方面表现优异，展示了其对于可扩展、可靠自动驾驶系统的实用价值。代码：https://github.com/Akina2001/DriveMVS.git。
英文原文摘要 1367 字符
Accurate metric depth is critical for autonomous driving perception and simulation, yet current approaches struggle to achieve high metric accuracy, multi-view and temporal consistency, and cross-domain generalization. To address these challenges, we present DriveMVS, a novel multi-view stereo framework that reconciles these competing objectives through two key insights: (1) Sparse but metrically accurate LiDAR observations can serve as geometric prompts to anchor depth estimation in absolute scale, and (2) deep fusion of diverse cues is essential for resolving ambiguities and enhancing robustness, while a spatio-temporal decoder ensures consistency across frames. Built upon these principles, DriveMVS embeds the LiDAR prompt in two ways: as a hard geometric prior that anchors the cost volume, and as soft feature-wise guidance fused by a triple-cue combiner. Regarding temporal consistency, DriveMVS employs a spatio-temporal decoder that jointly leverages geometric cues from the MVS cost volume and temporal context from neighboring frames. Experiments show that DriveMVS achieves state-of-the-art performance on multiple benchmarks, excelling in metric accuracy, temporal stability, and zero-shot cross-domain transfer, demonstrating its practical value for scalable, reliable autonomous driving systems. Code: https://github.com/Akina2001/DriveMVS.git.