Range-Agnostic Multi-View Depth Estimation With Keyframe Selection
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909082967867392 |
|---|---|
| author | Conti, Andrea Poggi, Matteo Cambareri, Valerio Mattoccia, Stefano |
| author_facet | Conti, Andrea Poggi, Matteo Cambareri, Valerio Mattoccia, Stefano |
| contents | Methods for 3D reconstruction from posed frames require prior knowledge about the scene metric range, usually to recover matching cues along the epipolar lines and narrow the search range. However, such prior might not be directly available or estimated inaccurately in real scenarios -- e.g., outdoor 3D reconstruction from video sequences -- therefore heavily hampering performance. In this paper, we focus on multi-view depth estimation without requiring prior knowledge about the metric range of the scene by proposing RAMDepth, an efficient and purely 2D framework that reverses the depth estimation and matching steps order. Moreover, we demonstrate the capability of our framework to provide rich insights about the quality of the views used for prediction. Additional material can be found on our project page https://andreaconti.github.io/projects/range_agnostic_multi_view_depth. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_14401 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Range-Agnostic Multi-View Depth Estimation With Keyframe Selection Conti, Andrea Poggi, Matteo Cambareri, Valerio Mattoccia, Stefano Computer Vision and Pattern Recognition Methods for 3D reconstruction from posed frames require prior knowledge about the scene metric range, usually to recover matching cues along the epipolar lines and narrow the search range. However, such prior might not be directly available or estimated inaccurately in real scenarios -- e.g., outdoor 3D reconstruction from video sequences -- therefore heavily hampering performance. In this paper, we focus on multi-view depth estimation without requiring prior knowledge about the metric range of the scene by proposing RAMDepth, an efficient and purely 2D framework that reverses the depth estimation and matching steps order. Moreover, we demonstrate the capability of our framework to provide rich insights about the quality of the views used for prediction. Additional material can be found on our project page https://andreaconti.github.io/projects/range_agnostic_multi_view_depth. |
| title | Range-Agnostic Multi-View Depth Estimation With Keyframe Selection |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2401.14401 |