LangGS-SLAM: Real-Time Language-Feature Gaussian Splatting SLAM
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908818291556352 |
|---|---|
| author | Ha, Seongbo Lee, Sibaek Kang, Kyungsu Choi, Joonyeol Tak, Seungjun Yu, Hyeonwoo |
| author_facet | Ha, Seongbo Lee, Sibaek Kang, Kyungsu Choi, Joonyeol Tak, Seungjun Yu, Hyeonwoo |
| contents | In this paper, we propose a RGB-D SLAM system that reconstructs a language-aligned dense feature field while sustaining low-latency tracking and mapping. First, we introduce a Top-K Rendering pipeline, a high-throughput and semantic-distortion-free method for efficiently rendering high-dimensional feature maps. To address the resulting semantic-geometric discrepancy and mitigate the memory consumption, we further design a multi-criteria map management strategy that prunes redundant or inconsistent Gaussians while preserving scene integrity. Finally, a hybrid field optimization framework jointly refines the geometric and semantic fields under real-time constraints by decoupling their optimization frequencies according to field characteristics. The proposed system achieves superior geometric fidelity compared to geometric-only baselines and comparable semantic fidelity to offline approaches while operating at 15 FPS. Our results demonstrate that online SLAM with dense, uncompressed language-aligned feature fields is both feasible and effective, bridging the gap between 3D perception and language-based reasoning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_06991 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | LangGS-SLAM: Real-Time Language-Feature Gaussian Splatting SLAM Ha, Seongbo Lee, Sibaek Kang, Kyungsu Choi, Joonyeol Tak, Seungjun Yu, Hyeonwoo Robotics Computer Vision and Pattern Recognition Image and Video Processing 68T45 In this paper, we propose a RGB-D SLAM system that reconstructs a language-aligned dense feature field while sustaining low-latency tracking and mapping. First, we introduce a Top-K Rendering pipeline, a high-throughput and semantic-distortion-free method for efficiently rendering high-dimensional feature maps. To address the resulting semantic-geometric discrepancy and mitigate the memory consumption, we further design a multi-criteria map management strategy that prunes redundant or inconsistent Gaussians while preserving scene integrity. Finally, a hybrid field optimization framework jointly refines the geometric and semantic fields under real-time constraints by decoupling their optimization frequencies according to field characteristics. The proposed system achieves superior geometric fidelity compared to geometric-only baselines and comparable semantic fidelity to offline approaches while operating at 15 FPS. Our results demonstrate that online SLAM with dense, uncompressed language-aligned feature fields is both feasible and effective, bridging the gap between 3D perception and language-based reasoning. |
| title | LangGS-SLAM: Real-Time Language-Feature Gaussian Splatting SLAM |
| topic | Robotics Computer Vision and Pattern Recognition Image and Video Processing 68T45 |
| url | https://arxiv.org/abs/2602.06991 |