FRACTAL: An Ultra-Large-Scale Aerial Lidar Dataset for 3D Semantic Segmentation of Diverse Landscapes

TL;DR

FRACTAL is a comprehensive, large-scale aerial Lidar dataset designed to improve 3D semantic segmentation across diverse landscapes, addressing limitations of existing benchmarks by offering high diversity and detailed annotations.

Contribution

The paper introduces FRACTAL, an ultra-large-scale Lidar dataset with diverse landscapes and high-quality labels, filling a critical gap for large-scale 3D land monitoring research.

Findings

High diversity and size of the dataset.

Baseline segmentation results demonstrate its utility.

Supports development of advanced 3D deep learning methods.

Abstract

Mapping agencies are increasingly adopting Aerial Lidar Scanning (ALS) as a new tool to map buildings and other above-ground structures. Processing ALS data at scale requires efficient point classification methods that perform well over highly diverse territories. Large annotated Lidar datasets are needed to evaluate these classification methods, however, current Lidar benchmarks have restricted scope and often cover a single urban area. To bridge this data gap, we introduce the FRench ALS Clouds from TArgeted Landscapes (FRACTAL) dataset: an ultra-large-scale aerial Lidar dataset made of 100,000 dense point clouds with high quality labels for 7 semantic classes and spanning 250 km$^2$. FRACTAL achieves high spatial and semantic diversity by explicitly sampling rare classes and challenging landscapes from five different regions of France. We describe the data collection, annotation, and curation process of the dataset. We provide baseline semantic segmentation results using a state of the art 3D point cloud classification model. FRACTAL aims to support the development of 3D deep learning approaches for large-scale land monitoring.