Multi-modal panoramic 3D outdoor datasets for place categorization

arXiv cs.RO / 4/16/2026

📰 NewsSignals & Early TrendsModels & Research

共有:

Key Points

The paper introduces two publicly available multi-modal panoramic 3D outdoor datasets (MPO) for semantic place categorization across six scene categories, including forest, coast, residential/urban areas, and indoor/outdoor parking lots.
The dense dataset contains 650 static panoramic scans with synchronized FARO laser point clouds (about 9,000,000 points) plus color and reflectance information.
The sparse dataset includes 34,200 real-time panoramic scans collected with a Velodyne LiDAR setup while driving, providing reflectance point clouds with about 70,000 points per scan.
Experiments compare multiple semantic place categorization approaches and report best accuracies of 96.42% for dense data and 89.67% for sparse data.
Data collection was performed in Fukuoka, Japan, and the authors provide dataset access links for researchers to benchmark and build on.

Abstract

We present two multi-modal panoramic 3D outdoor (MPO) datasets for semantic place categorization with six categories: forest, coast, residential area, urban area and indoor/outdoor parking lot. The first dataset consists of 650 static panoramic scans of dense (9,000,000 points) 3D color and reflectance point clouds obtained using a FARO laser scanner with synchronized color images. The second dataset consists of 34,200 real-time panoramic scans of sparse (70,000 points) 3D reflectance point clouds obtained using a Velodyne laser scanner while driving a car. The datasets were obtained in the city of Fukuoka, Japan and are publicly available in [1], [2]. In addition, we compare several approaches for semantic place categorization with best results of 96.42% (dense) and 89.67% (sparse).