Dense mapping from sparse visual odometry: a lightweight uncertainty-guaranteed depth completion method

doi:10.3389/frobt.2025.1644230

Dense mapping from sparse visual odometry: a lightweight uncertainty-guaranteed depth completion method

2025-10-01

0

OA

AI

DOI：10.3389/frobt.2025.1644230

Original

Original request for help

Save

Abstract

En 中文

IntroductionVisual odometry (VO) has been widely deployed on mobile robots for spatial perception. State-of-the-art VO offers robust localization; the maps it generates are often too sparse for downstream tasks due to insufffcient depth data. While depth completion methods can estimate dense depth from sparse data; the extreme sparsity and highly uneven distribution of depth signals in VO (∼ 0.15% of the pixels in the depth image available) poses signiffcant challenges.MethodsTo address this issue; we propose a lightweight Image-Guided Uncertainty-Aware Depth Completion Network (IU-DC) for completing sparse depth from VO. This network integrates color and spatial information into a normalized convolutional neural network to tackle the sparsity issue and simultaneously outputs dense depth and associated uncertainty. The estimated depth is uncertainty-aware; allowing for the filtering of outliers and ensuring precise spatial perception.ResultsThe superior performance of IU-DC compared to SOTA is validated across multiple open-source datasets in terms of depth and uncertainty estimation accuracy. In real-world mapping tasks; by integrating IU-DC with the mapping module; we achieve 50 × more reconstructed volumes and 78% coverage of the ground truth with twice the accuracy compared to SOTA; despite having only 0.6 M parameters (just 3% of the size of the SOTA).DiscussionOur code will be released at https://github.com/YangDL-BEIHANG/Dense-mapping-from-sparse-visual-odometry/tree/d5a11b4403b5ac2e9e0c3644b14b9711c2748bf9.

Dense mapping from sparse visual odometry: a lightweight uncertainty-guaranteed depth completion method

Abstract

Journal

Organization