NeRO: Neural Road Surface Reconstruction

TL;DR

This paper presents NeRO, a neural network framework using position encoding MLPs to accurately reconstruct road surfaces with height, color, and semantics from various data sources, suitable for autonomous driving applications.

Contribution

It introduces a novel position encoding MLP approach for road surface reconstruction that is compatible with multiple data sources and robust to semantic noise.

Findings

Compatible with vehicle camera, LiDAR, and SFM data

Robust to semantic noise and sparse labels

Fast training speed and effective for visualization

Abstract

Accurately reconstructing road surfaces is pivotal for various applications especially in autonomous driving. This paper introduces a position encoding Multi-Layer Perceptrons (MLPs) framework to reconstruct road surfaces, with input as world coordinates x and y, and output as height, color, and semantic information. The effectiveness of this method is demonstrated through its compatibility with a variety of road height sources like vehicle camera poses, LiDAR point clouds, and SFM point clouds, robust to the semantic noise of images like sparse labels and noise semantic prediction, and fast training speed, which indicates a promising application for rendering road surfaces with semantics, particularly in applications demanding visualization of road surface, 4D labeling, and semantic groupings.