City-scale Incremental Neural Mapping with Three-layer Sampling and Panoptic Representation

TL;DR

This paper introduces a city-scale neural mapping system that uses a three-layer sampling strategy and panoptic representation to improve the accuracy and detail of continual implicit mapping from sparse LiDAR data.

Contribution

It presents a novel city-scale continual neural mapping framework with a three-layer sampling method and panoptic representation, addressing geometric and instance-level modeling challenges.

Findings

Effective mapping of city-scale environments demonstrated on SemanticKITTI dataset.

Three-layer sampling improves geometric detail and local-global consistency.

Panoptic representation enhances instance-level mapping accuracy.

Abstract

Neural implicit representations are drawing a lot of attention from the robotics community recently, as they are expressive, continuous and compact. However, city-scale continual implicit dense mapping based on sparse LiDAR input is still an under-explored challenge. To this end, we successfully build a city-scale continual neural mapping system with a panoptic representation that consists of environment-level and instance-level modelling. Given a stream of sparse LiDAR point cloud, it maintains a dynamic generative model that maps 3D coordinates to signed distance field (SDF) values. To address the difficulty of representing geometric information at different levels in city-scale space, we propose a tailored three-layer sampling strategy to dynamically sample the global, local and near-surface domains. Meanwhile, to realize high fidelity mapping of instance under incomplete observation, category-specific prior is introduced to better model the geometric details. We evaluate on the public SemanticKITTI dataset and demonstrate the significance of the newly proposed three-layer sampling strategy and panoptic representation, using both quantitative and qualitative results. Codes and model will be publicly available.