English

Evaluating Robustness of Monocular Depth Estimation with Procedural Scene Perturbations

Computer Vision and Pattern Recognition 2025-07-04 v2

Abstract

Recent years have witnessed substantial progress on monocular depth estimation, particularly as measured by the success of large models on standard benchmarks. However, performance on standard benchmarks does not offer a complete assessment, because most evaluate accuracy but not robustness. In this work, we introduce PDE (Procedural Depth Evaluation), a new benchmark which enables systematic robustness evaluation. PDE uses procedural generation to create 3D scenes that test robustness to various controlled perturbations, including object, camera, material and lighting changes. Our analysis yields interesting findings on what perturbations are challenging for state-of-the-art depth models, which we hope will inform further research. Code and data are available at https://github.com/princeton-vl/proc-depth-eval.

Keywords

Cite

@article{arxiv.2507.00981,
  title  = {Evaluating Robustness of Monocular Depth Estimation with Procedural Scene Perturbations},
  author = {Jack Nugent and Siyang Wu and Zeyu Ma and Beining Han and Meenal Parakh and Abhishek Joshi and Lingjie Mei and Alexander Raistrick and Xinyuan Li and Jia Deng},
  journal= {arXiv preprint arXiv:2507.00981},
  year   = {2025}
}

Comments

Fixing display of figure on Safari browsers

R2 v1 2026-07-01T03:42:00.221Z