Large-Scale 3D Semantic Reconstruction for Automated Driving Vehicles with Adaptive Truncated Signed Distance Function

TL;DR

This paper introduces a novel large-scale 3D reconstruction system for autonomous vehicles that combines LiDAR and camera data using an adaptive signed distance function, improving model accuracy and semantic mapping.

Contribution

It proposes an Adaptive Truncated Signed Distance Function for better surface modeling from sparse LiDAR data and an optimal image patch selection for texturing, advancing outdoor 3D semantic reconstruction.

Findings

More accurate 3D models compared to state-of-the-art methods

Effective texturing from registered camera images

Promising semantic mapping results

Abstract

The Large-scale 3D reconstruction, texturing and semantic mapping are nowadays widely used for automated driving vehicles, virtual reality and automatic data generation. However, most approaches are developed for RGB-D cameras with colored dense point clouds and not suitable for large-scale outdoor environments using sparse LiDAR point clouds. Since a 3D surface can be usually observed from multiple camera images with different view poses, an optimal image patch selection for the texturing and an optimal semantic class estimation for the semantic mapping are still challenging. To address these problems, we propose a novel 3D reconstruction, texturing and semantic mapping system using LiDAR and camera sensors. An Adaptive Truncated Signed Distance Function is introduced to describe surfaces implicitly, which can deal with different LiDAR point sparsities and improve model quality. The from this implicit function extracted triangle mesh map is then textured from a series of registered camera images by applying an optimal image patch selection strategy. Besides that, a Markov Random Field-based data fusion approach is proposed to estimate the optimal semantic class for each triangle mesh. Our approach is evaluated on a synthetic dataset, the KITTI dataset and a dataset recorded with our experimental vehicle. The results show that the 3D models generated using our approach are more accurate in comparison to using other state-of-the-art approaches. The texturing and semantic mapping achieve also very promising results.