3D Gaussian Splatting aided Localization for Large and Complex Indoor-Environments
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911092621443072 |
|---|---|
| author | Ress, Vincent Meyer, Jonas Zhang, Wei Skuddis, David Soergel, Uwe Haala, Norbert |
| author_facet | Ress, Vincent Meyer, Jonas Zhang, Wei Skuddis, David Soergel, Uwe Haala, Norbert |
| contents | The field of visual localization has been researched for several decades and has meanwhile found many practical applications. Despite the strong progress in this field, there are still challenging situations in which established methods fail. We present an approach to significantly improve the accuracy and reliability of established visual localization methods by adding rendered images. In detail, we first use a modern visual SLAM approach that provides a 3D Gaussian Splatting (3DGS) based map to create reference data. We demonstrate that enriching reference data with images rendered from 3DGS at randomly sampled poses significantly improves the performance of both geometry-based visual localization and Scene Coordinate Regression (SCR) methods. Through comprehensive evaluation in a large industrial environment, we analyze the performance impact of incorporating these additional rendered views. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_13803 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | 3D Gaussian Splatting aided Localization for Large and Complex Indoor-Environments Ress, Vincent Meyer, Jonas Zhang, Wei Skuddis, David Soergel, Uwe Haala, Norbert Computer Vision and Pattern Recognition Robotics The field of visual localization has been researched for several decades and has meanwhile found many practical applications. Despite the strong progress in this field, there are still challenging situations in which established methods fail. We present an approach to significantly improve the accuracy and reliability of established visual localization methods by adding rendered images. In detail, we first use a modern visual SLAM approach that provides a 3D Gaussian Splatting (3DGS) based map to create reference data. We demonstrate that enriching reference data with images rendered from 3DGS at randomly sampled poses significantly improves the performance of both geometry-based visual localization and Scene Coordinate Regression (SCR) methods. Through comprehensive evaluation in a large industrial environment, we analyze the performance impact of incorporating these additional rendered views. |
| title | 3D Gaussian Splatting aided Localization for Large and Complex Indoor-Environments |
| topic | Computer Vision and Pattern Recognition Robotics |
| url | https://arxiv.org/abs/2502.13803 |