Segment Any 3D Gaussians
Proceedings of the AAAI Conference on Artificial Intelligence · 11.04.2025 · 39 (2) · 1971-1979
DOI: 10.1609/aaai.v39i2.32193 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 11.04.2025. Verfasst von Jiazhong Cen und 6 weiteren Autor:innen. Die Arbeit wurde bisher 117-mal zitiert (Crossref).
Abstract (Originalsprache)
This paper presents SAGA (Segment Any 3D GAussians), a highly efficient 3D promptable segmentation method based on 3D Gaussian Splatting (3D-GS). Given 2D visual prompts as input, SAGA can segment the corresponding 3D target represented by 3D Gaussians within 4 ms. This is achieved by attaching a scale-gated affinity feature to each 3D Gaussian to endow it a new property towards multi-granularity segmentation. Specifically, a scale-aware contrastive training strategy is proposed for the scale-gated affinity feature learning. It 1) distills the segmentation capability of the Segment Anything Model (SAM) from 2D masks into the affinity features and 2) employs a soft scale gate mechanism to deal with multi-granularity ambiguity in 3D segmentation through adjusting the magnitude of each feature channel according to a specified 3D physical scale. Evaluations demonstrate that SAGA achieves real-time multi-granularity segmentation with quality comparable to state-of-the-art methods. As one of the first methods addressing promptable segmentation in 3D-GS, the simplicity and effectiveness of SAGA pave the way for future advancements in this field.
Zitieren
APA
Cen, J., Fang, J., Yang, C., Xie, L., Zhang, X., Shen, W., & Tian, Q. (2025). Segment Any 3D Gaussians. Proceedings of the AAAI Conference on Artificial Intelligence, 39(2), 1971-1979. https://doi.org/10.1609/aaai.v39i2.32193
Vancouver
Cen J, Fang J, Yang C, Xie L, Zhang X, Shen W, et al. Segment Any 3D Gaussians. Proceedings of the AAAI Conference on Artificial Intelligence. 2025;39(2):1971-1979. doi:10.1609/aaai.v39i2.32193
BibTeX
@article{Cen2025,
author = {Cen, Jiazhong and Fang, Jiemin and Yang, Chen and Xie, Lingxi and Zhang, Xiaopeng and Shen, Wei and Tian, Qi},
title = {Segment Any 3D Gaussians},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2025},
volume = {39},
number = {2},
pages = {1971--1979},
doi = {10.1609/aaai.v39i2.32193}
}
EUVIMED