Deterministic Guided LiDAR Depth Map Completion

TL;DR

This paper introduces a non-deep learning approach for densifying sparse LiDAR depth maps using guided RGB images, involving artifact removal, superpixel segmentation, plane approximation, and interpolation, validated on the KITTI benchmark.

Contribution

It proposes a novel non-deep learning method for LiDAR depth map completion that outperforms existing methods on standard benchmarks.

Findings

Outperforms state-of-the-art non-deep learning methods.

Achieves competitive results compared to deep learning approaches.

Validated on KITTI depth completion benchmark.

Abstract

Accurate dense depth estimation is crucial for autonomous vehicles to analyze their environment. This paper presents a non-deep learning-based approach to densify a sparse LiDAR-based depth map using a guidance RGB image. To achieve this goal the RGB image is at first cleared from most of the camera-LiDAR misalignment artifacts. Afterward, it is over segmented and a plane for each superpixel is approximated. In the case a superpixel is not well represented by a plane, a plane is approximated for a convex hull of the most inlier. Finally, the pinhole camera model is used for the interpolation process and the remaining areas are interpolated. The evaluation of this work is executed using the KITTI depth completion benchmark, which validates the proposed work and shows that it outperforms the state-of-the-art non-deep learning-based methods, in addition to several deep learning-based methods.