Saved in:
Bibliographic Details
Main Authors: Sulzer, Raphael, Duan, Liuyun, Girard, Nicolas, Lafarge, Florent
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2505.15379
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910959460679680
author Sulzer, Raphael
Duan, Liuyun
Girard, Nicolas
Lafarge, Florent
author_facet Sulzer, Raphael
Duan, Liuyun
Girard, Nicolas
Lafarge, Florent
contents We present the P$^3$ dataset, a large-scale multimodal benchmark for building vectorization, constructed from aerial LiDAR point clouds, high-resolution aerial imagery, and vectorized 2D building outlines, collected across three continents. The dataset contains over 10 billion LiDAR points with decimeter-level accuracy and RGB images at a ground sampling distance of 25 centimeter. While many existing datasets primarily focus on the image modality, P$^3$ offers a complementary perspective by also incorporating dense 3D information. We demonstrate that LiDAR point clouds serve as a robust modality for predicting building polygons, both in hybrid and end-to-end learning frameworks. Moreover, fusing aerial LiDAR and imagery further improves accuracy and geometric quality of predicted polygons. The P$^3$ dataset is publicly available, along with code and pretrained weights of three state-of-the-art models for building polygon prediction at https://github.com/raphaelsulzer/PixelsPointsPolygons .
format Preprint
id arxiv_https___arxiv_org_abs_2505_15379
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle The P$^3$ dataset: Pixels, Points and Polygons for Multimodal Building Vectorization
Sulzer, Raphael
Duan, Liuyun
Girard, Nicolas
Lafarge, Florent
Computer Vision and Pattern Recognition
We present the P$^3$ dataset, a large-scale multimodal benchmark for building vectorization, constructed from aerial LiDAR point clouds, high-resolution aerial imagery, and vectorized 2D building outlines, collected across three continents. The dataset contains over 10 billion LiDAR points with decimeter-level accuracy and RGB images at a ground sampling distance of 25 centimeter. While many existing datasets primarily focus on the image modality, P$^3$ offers a complementary perspective by also incorporating dense 3D information. We demonstrate that LiDAR point clouds serve as a robust modality for predicting building polygons, both in hybrid and end-to-end learning frameworks. Moreover, fusing aerial LiDAR and imagery further improves accuracy and geometric quality of predicted polygons. The P$^3$ dataset is publicly available, along with code and pretrained weights of three state-of-the-art models for building polygon prediction at https://github.com/raphaelsulzer/PixelsPointsPolygons .
title The P$^3$ dataset: Pixels, Points and Polygons for Multimodal Building Vectorization
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.15379