Uncertainty Estimation for 3D Dense Prediction via Cross-Point Embeddings

Cai, Kaiwen, Lu, Chris Xiaoxuan and Huang, Xiaowei ORCID: 0000-0001-6267-0366 (2023) Uncertainty Estimation for 3D Dense Prediction via Cross-Point Embeddings. IEEE Robotics and Automation Letters, 8 (5). pp. 2558-2565.

PDF
_RAL_23__CUE.pdf - Author Accepted Manuscript
Download (8MB) | Preview

Official URL: http://dx.doi.org/10.1109/lra.2023.3256085

Abstract

Dense prediction tasks are common for 3D point clouds, but the uncertainties inherent in massive points and their embeddings have long been ignored. In this work, we present CUE, a novel uncertainty estimation method for dense prediction tasks in 3D point clouds. Inspired by metric learning, the key idea of CUE is to explore cross-point embeddings upon a conventional 3D dense prediction pipeline. Specifically, CUE involves building a probabilistic embedding model and then enforcing metric alignments of massive points in the embedding space. We also propose CUE+, which enhances CUE by explicitly modeling cross-point dependencies in the covariance matrix. We demonstrate that both CUE and CUE+ are generic and effective for uncertainty estimation in 3D point clouds with two different tasks: (1) in 3D geometric feature learning we for the first time obtain well-calibrated uncertainty, and (2) in semantic segmentation we reduce uncertainty's Expected Calibration Error of the state-of-the-arts by 16.5%. All uncertainties are estimated without compromising predictive performance.

Item Type:	Article
Additional Information:	(c) 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.
Uncontrolled Keywords:	Uncertainty, Point cloud compression, Three-dimensional displays, Task analysis, Probabilistic logic, Estimation, Semantic segmentation, Probabilistic inference, computer vision for automation, semantic scene understanding
Divisions:	Faculty of Science and Engineering > School of Electrical Engineering, Electronics and Computer Science
Depositing User:	Symplectic Admin
Date Deposited:	18 Apr 2023 08:41
Last Modified:	15 Mar 2024 13:59
DOI:	10.1109/lra.2023.3256085
Related URLs:	Author Publisher
URI:	https://livrepository.liverpool.ac.uk/id/eprint/3169649