Saved in:
Bibliographic Details
Main Authors: Liang, Siyun, Wang, Sen, Li, Kunyi, Niemeyer, Michael, Gasperini, Stefano, Lensch, Hendrik P. A., Navab, Nassir, Tombari, Federico
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.10231
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914266806747136
author Liang, Siyun
Wang, Sen
Li, Kunyi
Niemeyer, Michael
Gasperini, Stefano
Lensch, Hendrik P. A.
Navab, Nassir
Tombari, Federico
author_facet Liang, Siyun
Wang, Sen
Li, Kunyi
Niemeyer, Michael
Gasperini, Stefano
Lensch, Hendrik P. A.
Navab, Nassir
Tombari, Federico
contents 3D Gaussian Splatting has recently gained traction for its efficient training and real-time rendering. While its vanilla representation is mainly designed for view synthesis, recent works extended it to scene understanding with language features. However, storing additional high-dimensional features per Gaussian for semantic information is memory-intensive, which limits their ability to segment and interpret challenging scenes. To this end, we introduce SuperGSeg, a novel approach that fosters cohesive, context-aware hierarchical scene representation by disentangling segmentation and language field distillation. SuperGSeg first employs neural 3D Gaussians to learn geometry, instance and hierarchical segmentation features from multi-view images with the aid of off-the-shelf 2D masks. These features are then leveraged to create a sparse set of \acrlong{superg}s. \acrlong{superg}s facilitate the lifting and distillation of 2D language features into 3D space. They enable hierarchical scene understanding with high-dimensional language feature rendering at moderate GPU memory costs. Extensive experiments demonstrate that SuperGSeg achieves remarkable performance on both open-vocabulary object selection and semantic segmentation tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2412_10231
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SuperGSeg: Open-Vocabulary 3D Segmentation with Structured Super-Gaussians
Liang, Siyun
Wang, Sen
Li, Kunyi
Niemeyer, Michael
Gasperini, Stefano
Lensch, Hendrik P. A.
Navab, Nassir
Tombari, Federico
Computer Vision and Pattern Recognition
3D Gaussian Splatting has recently gained traction for its efficient training and real-time rendering. While its vanilla representation is mainly designed for view synthesis, recent works extended it to scene understanding with language features. However, storing additional high-dimensional features per Gaussian for semantic information is memory-intensive, which limits their ability to segment and interpret challenging scenes. To this end, we introduce SuperGSeg, a novel approach that fosters cohesive, context-aware hierarchical scene representation by disentangling segmentation and language field distillation. SuperGSeg first employs neural 3D Gaussians to learn geometry, instance and hierarchical segmentation features from multi-view images with the aid of off-the-shelf 2D masks. These features are then leveraged to create a sparse set of \acrlong{superg}s. \acrlong{superg}s facilitate the lifting and distillation of 2D language features into 3D space. They enable hierarchical scene understanding with high-dimensional language feature rendering at moderate GPU memory costs. Extensive experiments demonstrate that SuperGSeg achieves remarkable performance on both open-vocabulary object selection and semantic segmentation tasks.
title SuperGSeg: Open-Vocabulary 3D Segmentation with Structured Super-Gaussians
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.10231