Multimodal sensor dataset from vehicle-mounted mobile mapping system for comprehensive urban scenes

Area of Science:

• Geomatics and Geospatial Science
• Robotics and Autonomous Systems
• Computer Vision and Machine Learning

Background:

• Mobile mapping offers superior time efficiency over traditional methods, crucial for applications like high-definition maps and smart cities.
• Existing mobile mapping datasets often rely on portable platforms, limiting research in large-scale autonomous driving applications.
• There is a need for comprehensive, vehicle-mounted mobile mapping datasets to advance urban scene understanding and autonomous systems.

Purpose of the Study:

• To introduce the multimodal sensor dataset from a vehicle-mounted mobile mapping system for comprehensive urban scenes (MSD-VMMS-HK).
• To provide a rich, high-precision, large-scale dataset for diverse urban environments, including challenging scenarios.
• To establish a benchmark for validating point cloud and image processing techniques in urban mapping.

Main Methods:

• Utilized a vehicle-mounted mobile mapping system equipped with high-precision (millimeter-level) LiDAR, panoramic cameras, and GNSS/INS.
• Collected data across a wide variety of representative urban scenarios in Hong Kong, including tunnels, canyons, and coastal roads.
• Developed a multimodal dataset enabling precise spatial and visual data fusion for urban scene analysis.

Main Results:

• The MSD-VMMS-HK dataset provides high-precision, large-scale multimodal sensor information for complex urban environments.
• It is the first urban-level dataset offering high-precision references for validating point cloud and image processing algorithms.
• Demonstrated dataset applications in accurate urban canyon mapping, infrastructure management, and change detection.

Conclusions:

• The MSD-VMMS-HK dataset significantly advances research in large-scale mobile mapping and autonomous driving.
• This comprehensive dataset serves as a valuable resource for the academic community in developing and validating urban mapping technologies.
• Facilitates the creation of digital twins, smart city infrastructure, and improved autonomous navigation systems.