Privacy-Preserving Depth-Only Open-Vocabulary 3D Semantic Segmentation Via Uncertainty-Guided Test-Time Optimization
Updated
Updated · arxiv.org · Jul 1
Privacy-Preserving Depth-Only Open-Vocabulary 3D Semantic Segmentation Via Uncertainty-Guided Test-Time Optimization
1 articles · Updated · arxiv.org · Jul 1
Summary
Researchers have introduced UTTO, a framework for privacy-preserving, depth-only open-vocabulary 3D semantic segmentation that requires no RGB images.
UTTO uses uncertainty-guided test-time optimization to improve prediction reliability, leveraging only depth-derived geometry and semantic priors from foundation models.
This approach addresses privacy concerns in indoor environments and narrows the performance gap between RGB-D and geometry-only 3D scene understanding.