LighthouseGS: Indoor Structure-aware 3D Gaussian Splatting for Panorama-Style Mobile Captures
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910018255716352 |
|---|---|
| author | Han, Seungoh Jang, Jaehoon Kim, Hyunsu Surh, Jaeheung Kwak, Junhyung Ha, Hyowon Joo, Kyungdon |
| author_facet | Han, Seungoh Jang, Jaehoon Kim, Hyunsu Surh, Jaeheung Kwak, Junhyung Ha, Hyowon Joo, Kyungdon |
| contents | We introduce LighthouseGS, a practical novel view synthesis framework based on 3D Gaussian Splatting that utilizes simple panorama-style captures from a single mobile device. While convenient, this rotation-dominant motion and narrow baseline make accurate camera pose and 3D point estimation challenging, especially in textureless indoor scenes. To address these challenges, LighthouseGS leverages rough geometric priors, such as mobile device camera poses and monocular depth estimation, and utilizes indoor planar structures. Specifically, we propose a new initialization method called plane scaffold assembly to generate consistent 3D points on these structures, followed by a stable pruning strategy to enhance geometry and optimization stability. Additionally, we present geometric and photometric corrections to resolve inconsistencies from motion drift and auto-exposure in mobile devices. Tested on real and synthetic indoor scenes, LighthouseGS delivers photorealistic rendering, outperforming state-of-the-art methods and enabling applications like panoramic view synthesis and object placement. Project page: https://vision3d-lab.github.io/lighthousegs/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_06109 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | LighthouseGS: Indoor Structure-aware 3D Gaussian Splatting for Panorama-Style Mobile Captures Han, Seungoh Jang, Jaehoon Kim, Hyunsu Surh, Jaeheung Kwak, Junhyung Ha, Hyowon Joo, Kyungdon Graphics Artificial Intelligence Computer Vision and Pattern Recognition We introduce LighthouseGS, a practical novel view synthesis framework based on 3D Gaussian Splatting that utilizes simple panorama-style captures from a single mobile device. While convenient, this rotation-dominant motion and narrow baseline make accurate camera pose and 3D point estimation challenging, especially in textureless indoor scenes. To address these challenges, LighthouseGS leverages rough geometric priors, such as mobile device camera poses and monocular depth estimation, and utilizes indoor planar structures. Specifically, we propose a new initialization method called plane scaffold assembly to generate consistent 3D points on these structures, followed by a stable pruning strategy to enhance geometry and optimization stability. Additionally, we present geometric and photometric corrections to resolve inconsistencies from motion drift and auto-exposure in mobile devices. Tested on real and synthetic indoor scenes, LighthouseGS delivers photorealistic rendering, outperforming state-of-the-art methods and enabling applications like panoramic view synthesis and object placement. Project page: https://vision3d-lab.github.io/lighthousegs/ |
| title | LighthouseGS: Indoor Structure-aware 3D Gaussian Splatting for Panorama-Style Mobile Captures |
| topic | Graphics Artificial Intelligence Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2507.06109 |