The Unreasonable Effectiveness of Pre-Trained Features for Camera Pose Refinement
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917641470345216 |
|---|---|
| author | Trivigno, Gabriele Masone, Carlo Caputo, Barbara Sattler, Torsten |
| author_facet | Trivigno, Gabriele Masone, Carlo Caputo, Barbara Sattler, Torsten |
| contents | Pose refinement is an interesting and practically relevant research direction. Pose refinement can be used to (1) obtain a more accurate pose estimate from an initial prior (e.g., from retrieval), (2) as pre-processing, i.e., to provide a better starting point to a more expensive pose estimator, (3) as post-processing of a more accurate localizer. Existing approaches focus on learning features / scene representations for the pose refinement task. This involves training an implicit scene representation or learning features while optimizing a camera pose-based loss. A natural question is whether training specific features / representations is truly necessary or whether similar results can be already achieved with more generic features. In this work, we present a simple approach that combines pre-trained features with a particle filter and a renderable representation of the scene. Despite its simplicity, it achieves state-of-the-art results, demonstrating that one can easily build a pose refiner without the need for specific training. The code is at https://github.com/ga1i13o/mcloc_poseref |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_10438 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | The Unreasonable Effectiveness of Pre-Trained Features for Camera Pose Refinement Trivigno, Gabriele Masone, Carlo Caputo, Barbara Sattler, Torsten Computer Vision and Pattern Recognition Pose refinement is an interesting and practically relevant research direction. Pose refinement can be used to (1) obtain a more accurate pose estimate from an initial prior (e.g., from retrieval), (2) as pre-processing, i.e., to provide a better starting point to a more expensive pose estimator, (3) as post-processing of a more accurate localizer. Existing approaches focus on learning features / scene representations for the pose refinement task. This involves training an implicit scene representation or learning features while optimizing a camera pose-based loss. A natural question is whether training specific features / representations is truly necessary or whether similar results can be already achieved with more generic features. In this work, we present a simple approach that combines pre-trained features with a particle filter and a renderable representation of the scene. Despite its simplicity, it achieves state-of-the-art results, demonstrating that one can easily build a pose refiner without the need for specific training. The code is at https://github.com/ga1i13o/mcloc_poseref |
| title | The Unreasonable Effectiveness of Pre-Trained Features for Camera Pose Refinement |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2404.10438 |