AI & ML
Depth-Only Open-Vocabulary 3D Semantic Segmentation For Privacy-Preserving Robotic Applications
arXiv cs.RO0 views
arXiv:2607.00978v3 Announce Type: replace-cross Abstract: Privacy-preserving perception is increasingly important for robotic systems operating in real-world indoor environments, yet it remains underexplored in open-vocabulary 3D semantic segmentation. We study this problem under an RGB-prohibited deployment setting motivated by scene-specific visual information disclosure, where real RGB observations are unavailable during scene acquisition and fusion. To reflect this deployment constraint on existing 3D datasets, we adopt a stricter depth-only evaluation protocol that re-runs scene fusion without RGB and exposes only the resulting depth-derived geometry to the segmentation pipeline. This constraint removes appearance cues that are often critical for open-vocabulary recognition, making depth-only predictions more uncertain and less reliable. To address this challenge, we propose UTTO, a model-agnostic uncertainty-guided test-time optimization framework that uses structured predictive uncertainty as a reliability signal to refine predictions from frozen open-vocabulary 3D backbones. Experiments across ScanNet and Matterport3D demonstrate consistent improvements over multiple depth-only backbones. Privacy recoverability analyses and a real-robot semantic goal grounding case study further support the proposed privacy-constrained setting and applicability.
Source:https://arxiv.org/abs/2607.00978PreviousCurvature-aware 3D length estimation of greenhouse cucumbers using RGB-D imaging and cubic spline arc-length integrationNextTowards end-to-end optimization in multimaterial 3D printing
Related posts
We shipped a free motion design editor for iPhone: KeyCut Mobile
0dev.to Design
What College Didn’t Teach Me About Getting My First Tech Job
0dev.to Discuss

What Makes On-Demand Apps Technology Important for Business Expansion
0dev.to Discuss
Forward a cold email to a colleague and the follow-ups keep coming. I read 4 help centers to find out why.
0dev.to Discuss
AI-trained 5-tonne John Deere loader performs world-record acrobatics on two wheels
0dev.to Machine Learning
How Netflix and Spotify Know What You Want Next
0dev.to Machine Learning