UQ-Loc: Uncertainty-Aware LiDAR Scene Coordinate Regression
2026-08-06 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors developed UQ-Loc, an improved method for locating a camera or device in 3D space using LiDAR data. Unlike previous methods that gave only fixed position estimates, their model also predicts uncertainty about those estimates, helping the system know how confident it is. They trained their model with a special loss function and used uncertainty information to improve the localization process. Tests showed that their approach is more accurate and provides reliable uncertainty measurements.
LiDARScene Coordinate Regression6-DoF LocalizationAleatoric UncertaintyAnisotropic Gaussian CovarianceNegative Log-LikelihoodSpatial Smoothness RegularizerMahalanobis DistanceExpected Calibration Error
Authors
Jacek Komorowski
Abstract
LiDAR-based Scene Coordinate Regression (SCR) maps point clouds directly to 3D scene coordinates, enabling precise 6-DoF localisation without explicit map retrieval. However, existing methods produce deterministic predictions, discarding aleatoric uncertainty that could improve robustness and downstream decision-making. We present UQ-Loc, which extends the LightLoc architecture with an anisotropic Gaussian covariance head that predicts a full 3x3 positive-definite covariance matrix per voxel. Training uses a Negative Log-Likelihood (NLL) loss augmented with a kNN-based spatial smoothness regulariser, while inference employs a modified SC2-PCR solver with uncertainty-weighted seed scoring and a Mahalanobis-distance inlier test. We adopt Expected Calibration Error (ECE) as a principled metric for evaluating the quality of the predicted uncertainty. Experiments demonstrate that UQ-Loc achieves consistent improvement in 6-DoF localization accuracy while producing well-calibrated covariances.