Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2510.19814 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866918205197385728 |
|---|---|
| author | Wu, Siyang Nugent, Jack Yang, Willow Deng, Jia |
| author_facet | Wu, Siyang Nugent, Jack Yang, Willow Deng, Jia |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_19814 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Toward A Better Understanding of Monocular Depth Evaluation Wu, Siyang Nugent, Jack Yang, Willow Deng, Jia Computer Vision and Pattern Recognition 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. |
| title | Toward A Better Understanding of Monocular Depth Evaluation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2510.19814 |