No Adaptation Without Observation: Observability-Constrained Test-Time Prompt Tuning for LiDAR Semantic Segmentation

June 29, 2026 ยท Grace Period ยท ๐Ÿ› IROS 2026

โณ Grace Period
This paper is less than 90 days old. We give authors time to release their code before passing judgment.
Authors Linlian Jiang, Wentao Ju, Sadman Rakib Pinon, Jianwei Xian, Zhixiang Chi, Xinxin Zuo, Yang Wang arXiv ID 2606.30937 Category cs.CV: Computer Vision Citations 0 Venue IROS 2026
Abstract
LiDAR semantic segmentation often degrades under real-world deployment due to evolving sensing conditions, while collecting new annotations for retraining is impractical. Test-time adaptation (TTA) updates model parameters online using pseudo-label supervision, but directly applying standard TTA strategies to LiDAR data is challenging. Because pseudo-label reliability is spatially heteroscedastic under range-dependent sparsity and occlusion, uniform updates on globally shared parameters can inject unstable gradients and destabilize adaptation. We propose a geometry-constrained test-time prompt tuning framework for LiDAR semantic segmentation. Our method estimates per-location sensing reliability from depth-consistent beam terminations and neighborhood support, and uses it to reweight spatial supervision. Adaptation is confined to lightweight prompt adapters inserted into a frozen backbone, with spatial gating to prevent unreliable regions from perturbing globally shared representations. A temporally smoothed prototype alignment strategy further stabilizes online updates by accumulating reliable semantic evidence over time. Experiments on standard LiDAR benchmarks demonstrate improved adaptation stability and segmentation performance under deployment variations without additional annotations.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

๐Ÿ“œ Similar Papers

In the same crypt โ€” Computer Vision

๐ŸŒ… ๐ŸŒ… Old Age

Fast R-CNN

Ross Girshick

cs.CV ๐Ÿ› ICCV ๐Ÿ“š 27.7K cites 11 years ago