When geospatial big data meets high performance computing in 3D GIS

Authors: Tianyang Chen*, University of North Carolina - Charlotte, Wenwu Tang, Advisor
Topics: Geographic Information Science and Systems, Spatial Analysis & Modeling
Keywords: 3D GIS, structure from motion, high performance computing
Session Type: Paper
Day: 4/5/2019
Start / End Time: 8:00 AM / 9:40 AM
Room: Roosevelt 3, Marriott, Exhibition Level
Presentation File: No File Uploaded


3D GIS is one of the cutting-edge topics in geography, flooding many subjects, such as urban planning, environmental assessment, and disaster management. Structure from motion (SfM), a state-of-art technique for 3D reconstruction based on photogrammetry, has warmed up in geospatial data acquisition and analytics of 3D GIS. However, SfM faces a geospatial big data challenge that dramatically slows down the associated processing and analytics. Most of the existing solutions are to make a compromise to the computing speed by sacrificing the accuracy of the modeling. High-performance computing driven by cyberinfrastructure could be a potential solution in facing such massive computing. Coupling with high-performance computing, SfM would release more ability in 3D modeling without making the compromise between computing and accuracy. Therefore, we developed an approach that integrates structure from motion and high-performance computing to bridge the gaps between the demand for highly accurate modeling in 3D GIS and the low efficiency of SfM. A case study of 3D geospatial data modeling related to hydraulic structures such as bridges is conducted to evaluate the performance of this approach.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login