# 3D Surface Reconstruction from Voxel-based Lidar Data

## TL;DR

This paper presents a novel 3D surface reconstruction algorithm from voxel-based Lidar data using an adaptive TSDF approach, improving mesh accuracy and density for autonomous vehicle navigation.

## Contribution

Introduces an adaptive neighborhood kernel based on Gaussian confidence for TSDF voxel representation, enhancing surface reconstruction quality.

## Key findings

- Effective on synthetic and real datasets
- Outperforms existing surface reconstruction methods
- Balances mesh density and accuracy

## Abstract

To achieve fully autonomous navigation, vehicles need to compute an accurate model of their direct surrounding. In this paper, a 3D surface reconstruction algorithm from heterogeneous density 3D data is presented. The proposed method is based on a TSDF voxel-based representation, where an adaptive neighborhood kernel sourced on a Gaussian confidence evaluation is introduced. This enables to keep a good trade-off between the density of the reconstructed mesh and its accuracy. Experimental evaluations carried on both synthetic (CARLA) and real (KITTI) 3D data show a good performance compared to a state of the art method used for surface reconstruction.