LabelGS: Label-Aware 3D Gaussian Splatting for 3D Scene Segmentation
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_ | 1866914009014337536 |
|---|---|
| author | Zhang, Yupeng Zheng, Dezhi Lu, Ping Zhang, Han Wang, Lei xiang, Liping Luo, Cheng Deng, Kaijun Fu, Xiaowen Shen, Linlin Wang, Jinbao |
| author_facet | Zhang, Yupeng Zheng, Dezhi Lu, Ping Zhang, Han Wang, Lei xiang, Liping Luo, Cheng Deng, Kaijun Fu, Xiaowen Shen, Linlin Wang, Jinbao |
| contents | 3D Gaussian Splatting (3DGS) has emerged as a novel explicit representation for 3D scenes, offering both high-fidelity reconstruction and efficient rendering. However, 3DGS lacks 3D segmentation ability, which limits its applicability in tasks that require scene understanding. The identification and isolating of specific object components is crucial. To address this limitation, we propose Label-aware 3D Gaussian Splatting (LabelGS), a method that augments the Gaussian representation with object label.LabelGS introduces cross-view consistent semantic masks for 3D Gaussians and employs a novel Occlusion Analysis Model to avoid overfitting occlusion during optimization, Main Gaussian Labeling model to lift 2D semantic prior to 3D Gaussian and Gaussian Projection Filter to avoid Gaussian label conflict. Our approach achieves effective decoupling of Gaussian representations and refines the 3DGS optimization process through a random region sampling strategy, significantly improving efficiency. Extensive experiments demonstrate that LabelGS outperforms previous state-of-the-art methods, including Feature-3DGS, in the 3D scene segmentation task. Notably, LabelGS achieves a remarkable 22X speedup in training compared to Feature-3DGS, at a resolution of 1440X1080. Our code will be at https://github.com/garrisonz/LabelGS. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_19699 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | LabelGS: Label-Aware 3D Gaussian Splatting for 3D Scene Segmentation Zhang, Yupeng Zheng, Dezhi Lu, Ping Zhang, Han Wang, Lei xiang, Liping Luo, Cheng Deng, Kaijun Fu, Xiaowen Shen, Linlin Wang, Jinbao Computer Vision and Pattern Recognition 3D Gaussian Splatting (3DGS) has emerged as a novel explicit representation for 3D scenes, offering both high-fidelity reconstruction and efficient rendering. However, 3DGS lacks 3D segmentation ability, which limits its applicability in tasks that require scene understanding. The identification and isolating of specific object components is crucial. To address this limitation, we propose Label-aware 3D Gaussian Splatting (LabelGS), a method that augments the Gaussian representation with object label.LabelGS introduces cross-view consistent semantic masks for 3D Gaussians and employs a novel Occlusion Analysis Model to avoid overfitting occlusion during optimization, Main Gaussian Labeling model to lift 2D semantic prior to 3D Gaussian and Gaussian Projection Filter to avoid Gaussian label conflict. Our approach achieves effective decoupling of Gaussian representations and refines the 3DGS optimization process through a random region sampling strategy, significantly improving efficiency. Extensive experiments demonstrate that LabelGS outperforms previous state-of-the-art methods, including Feature-3DGS, in the 3D scene segmentation task. Notably, LabelGS achieves a remarkable 22X speedup in training compared to Feature-3DGS, at a resolution of 1440X1080. Our code will be at https://github.com/garrisonz/LabelGS. |
| title | LabelGS: Label-Aware 3D Gaussian Splatting for 3D Scene Segmentation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2508.19699 |