Rethinking End-to-End 2D to 3D Scene Segmentation in Gaussian Splatting
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866913742766211072 |
|---|---|
| author | Zhu, Runsong Qiu, Shi Liu, Zhengzhe Hui, Ka-Hei Wu, Qianyi Heng, Pheng-Ann Fu, Chi-Wing |
| author_facet | Zhu, Runsong Qiu, Shi Liu, Zhengzhe Hui, Ka-Hei Wu, Qianyi Heng, Pheng-Ann Fu, Chi-Wing |
| contents | Lifting multi-view 2D instance segmentation to a radiance field has proven to be effective to enhance 3D understanding. Existing methods rely on direct matching for end-to-end lifting, yielding inferior results; or employ a two-stage solution constrained by complex pre- or post-processing. In this work, we design a new end-to-end object-aware lifting approach, named Unified-Lift that provides accurate 3D segmentation based on the 3D Gaussian representation. To start, we augment each Gaussian point with an additional Gaussian-level feature learned using a contrastive loss to encode instance information. Importantly, we introduce a learnable object-level codebook to account for individual objects in the scene for an explicit object-level understanding and associate the encoded object-level features with the Gaussian-level point features for segmentation predictions. While promising, achieving effective codebook learning is non-trivial and a naive solution leads to degraded performance. Therefore, we formulate the association learning module and the noisy label filtering module for effective and robust codebook learning. We conduct experiments on three benchmarks: LERF-Masked, Replica, and Messy Rooms datasets. Both qualitative and quantitative results manifest that our Unified-Lift clearly outperforms existing methods in terms of segmentation quality and time efficiency. The code is publicly available at \href{https://github.com/Runsong123/Unified-Lift}{https://github.com/Runsong123/Unified-Lift}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_14029 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Rethinking End-to-End 2D to 3D Scene Segmentation in Gaussian Splatting Zhu, Runsong Qiu, Shi Liu, Zhengzhe Hui, Ka-Hei Wu, Qianyi Heng, Pheng-Ann Fu, Chi-Wing Computer Vision and Pattern Recognition Lifting multi-view 2D instance segmentation to a radiance field has proven to be effective to enhance 3D understanding. Existing methods rely on direct matching for end-to-end lifting, yielding inferior results; or employ a two-stage solution constrained by complex pre- or post-processing. In this work, we design a new end-to-end object-aware lifting approach, named Unified-Lift that provides accurate 3D segmentation based on the 3D Gaussian representation. To start, we augment each Gaussian point with an additional Gaussian-level feature learned using a contrastive loss to encode instance information. Importantly, we introduce a learnable object-level codebook to account for individual objects in the scene for an explicit object-level understanding and associate the encoded object-level features with the Gaussian-level point features for segmentation predictions. While promising, achieving effective codebook learning is non-trivial and a naive solution leads to degraded performance. Therefore, we formulate the association learning module and the noisy label filtering module for effective and robust codebook learning. We conduct experiments on three benchmarks: LERF-Masked, Replica, and Messy Rooms datasets. Both qualitative and quantitative results manifest that our Unified-Lift clearly outperforms existing methods in terms of segmentation quality and time efficiency. The code is publicly available at \href{https://github.com/Runsong123/Unified-Lift}{https://github.com/Runsong123/Unified-Lift}. |
| title | Rethinking End-to-End 2D to 3D Scene Segmentation in Gaussian Splatting |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2503.14029 |