Abstract
Open-vocabulary 3D segmentation localizes objects from free-form text queries, but remains challenging in real image sequences: incomplete or noisy 2D supervision destabilizes multi-view identity assignment, while full-scene semantic learning weakens object-level discriminability. We introduce Hybrid Gaussians, a unified 3D representation jointly modeling object association and language-aligned semantics. Its Multi-View Object Association mechanism combines Observation Fusion and Semantic Contrastive Learning to improve identity consistency and semantic discrimination. Boundary Reconstruction Optimization further refines local boundary structure to improve contour quality. Experiments on LERF and 3D-OVS demonstrate strong quantitative and qualitative performance. Our method achieves 59.1\% mIoU on LERF, yielding a 13.4\% relative gain over the baseline. Project page: https://nora202.github.io/hybridgaussians.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Qiu, X., Sun, Y., Zhang, T., Xia, Y., & Long, Y. (2026). Hybrid Gaussians for Robust Open-Vocabulary 3D Segmentation with Multi-View Object Association and Boundary Refinement. https://omanscience.com/en/articles/hybrid-gaussians-for-robust-open-vocabulary-3d-segmentation-with-multi-view-object-association-and-boundary-refinement
MLA 9
Qiu, Xueqi, et al. "Hybrid Gaussians for Robust Open-Vocabulary 3D Segmentation with Multi-View Object Association and Boundary Refinement." https://omanscience.com/en/articles/hybrid-gaussians-for-robust-open-vocabulary-3d-segmentation-with-multi-view-object-association-and-boundary-refinement.
Chicago (author–date)
Qiu, Xueqi, Yueming Sun, Tianyu Zhang, Yuxuan Xia, and Yang Long. 2026. "Hybrid Gaussians for Robust Open-Vocabulary 3D Segmentation with Multi-View Object Association and Boundary Refinement." https://omanscience.com/en/articles/hybrid-gaussians-for-robust-open-vocabulary-3d-segmentation-with-multi-view-object-association-and-boundary-refinement.
Harvard
Qiu, X., Sun, Y., Zhang, T., Xia, Y. and Long, Y. (2026) 'Hybrid Gaussians for Robust Open-Vocabulary 3D Segmentation with Multi-View Object Association and Boundary Refinement', Available at: https://omanscience.com/en/articles/hybrid-gaussians-for-robust-open-vocabulary-3d-segmentation-with-multi-view-object-association-and-boundary-refinement.
Vancouver
Qiu X, Sun Y, Zhang T, Xia Y, Long Y. Hybrid Gaussians for Robust Open-Vocabulary 3D Segmentation with Multi-View Object Association and Boundary Refinement. https://omanscience.com/en/articles/hybrid-gaussians-for-robust-open-vocabulary-3d-segmentation-with-multi-view-object-association-and-boundary-refinement
IEEE
X. Qiu, Y. Sun, T. Zhang, Y. Xia, and Y. Long, "Hybrid Gaussians for Robust Open-Vocabulary 3D Segmentation with Multi-View Object Association and Boundary Refinement," https://omanscience.com/en/articles/hybrid-gaussians-for-robust-open-vocabulary-3d-segmentation-with-multi-view-object-association-and-boundary-refinement.