Accession Number:

AD1058665

Title:

3D Modeling of Urban Sites from Point Clouds

Descriptive Note:

Technical Report,12 Aug 2013,11 Dec 2016

Corporate Author:

University of Southern California Los Angeles United States

Personal Author(s):

• You, Suya

Report Date:

2017-05-03

Pagination or Media Count:

18.0

Abstract:

This project focuses on the developments of advanced techniques for processing 3D point cloud data specially targeted for 3D urban modeling. Current work on urban modeling has focused on either isolated objects or large-scale structures, not the wide diversity of small-scale objects occurring in urban scenes. The ultimate objective of this research is to pursue new techniques and solutions that extend the range of object classes that can be rapidly discovered, recognized, extracted and modeled from point-cloud data.

Descriptors:

• deep learning
• recognition
• classification
• complex systems
• machine learning
• pattern recognition
• artificial neural networks
• detection
• change detection
• computer vision
• point clouds
• urban areas

Subject Categories:

• Cybernetics
• Theoretical Mathematics
• Target Direction, Range and Position Finding

Distribution Statement:

APPROVED FOR PUBLIC RELEASE