提出新评价方法，让单目深度估计更贴近人眼判断。
Toward A Better Understanding of Monocular Depth Evaluation
- 基于相对表面法向设计新指标，捕捉曲面细节变化。
- 发现现有指标对曲率扰动不敏感，易忽略表面凹凸差异。
- 提供可视化工具与组合指标方法，适合深度评估研究者使用。
单目深度估计进展迅速，但其评价体系尚未标准化，现有评估指标种类繁多且理解不足。本文通过定量分析各类指标对真实值扰动的敏感性，特别关注其与人类判断的一致性。结果表明，现有指标对曲率扰动（如将平滑表面变为粗糙）极为不敏感。为此，我们提出一种基于相对表面法向的新指标，并配套开发深度可视化工具及可组合的评估方法，以提升指标与人类感知的对齐度。代码与数据已开源：https://github.com/princeton-vl/evalmde。
原文摘要 · Abstract (English)
Monocular depth estimation is an important task with rapid progress, but how to evaluate it is not fully resolved, as evidenced by a lack of standardization in existing literature and a large selection of evaluation metrics whose trade-offs and behaviors are not fully understood. This paper contributes a novel, quantitative analysis of existing metrics in terms of their sensitivity to various types of perturbations of ground truth, emphasizing comparison to human judgment. Our analysis reveals that existing metrics are severely under-sensitive to curvature perturbation such as making smooth surfaces bumpy. To remedy this, we introduce a new metric based on relative surface normals, along with new depth visualization tools and a principled method to create composite metrics with better human alignment. Code and data are available at: https://github.com/princeton-vl/evalmde.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。