度量深度的点石成金
The Midas Touch for Metric Depth
把现成的相对深度预测结果用极稀疏的 3D 点（如 LiDAR 稀疏点、SfM 点）提升为度量深度，分段稀疏图优化加不连续感知测地代价的像素级细化，可直接作为后处理模块接在任意单目深度模型后面，用于需要真实尺度的下游 3D 任务。
原文依据 converts relative depth into metric depth using only extremely sparse 3D data
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论近期进展显著提升了相对深度估计的跨场景泛化能力，但其实际可用性仍受限于缺乏度量尺度、局部不一致以及计算效率低下。为解决这些问题，本文提出 Midas Touch for Depth（MTD），一种仅利用极稀疏 3D 数据即可将相对深度转换为度量深度的、具有数学可解释性的方法。为消除局部尺度不一致，该方法通过稀疏图优化实施分段恢复策略，随后采用基于不连续感知测地代价的像素级细化策略。MTD 展现出强泛化能力，相较以往的深度补全与深度估计方法取得了显著的精度提升。此外，其轻量、即插即用的设计便于在多种下游 3D 任务上部署与集成。
英文原文摘要 892 字符
Recent advances have markedly improved the cross-scene generalization of relative depth estimation, yet its practical applicability remains limited by the absence of metric scale, local inconsistencies, and low computational efficiency. To address these issues, we present Midas Touch for Depth (MTD), a mathematically interpretable approach that converts relative depth into metric depth using only extremely sparse 3D data. To eliminate local scale inconsistencies, it applies a segment-wise recovery strategy via sparse graph optimization, followed by a pixel-wise refinement strategy using a discontinuity-aware geodesic cost. MTD exhibits strong generalization and achieves substantial accuracy improvements over previous depth completion and depth estimation methods. Moreover, its lightweight, plug-and-play design facilitates deployment and integration on diverse downstream 3D tasks.