Share Email Print
cover

Proceedings Paper

Improving object detection in 2D images using a 3D world model
Author(s): Herbert E. M. Viggh; Peter L. Cho; Nicholas Armstrong-Crews; Myra Nam; Danelle C. Shah; Geoffrey E. Brown
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

A mobile robot operating in a netcentric environment can utilize offboard resources on the network to improve its local perception. One such offboard resource is a world model built and maintained by other sensor systems. In this paper we present results from research into improving the performance of Deformable Parts Model object detection algorithms by using an offboard 3D world model. Experiments were run for detecting both people and cars in 2D photographs taken in an urban environment. After generating candidate object detections, a 3D world model built from airborne Light Detection and Ranging (LIDAR) and aerial photographs was used to filter out false alarm using several types of geometric reasoning. Comparison of the baseline detection performance to the performance after false alarm filtering showed a significant decrease in false alarms for a given probability of detection.

Paper Details

Date Published: 22 May 2014
PDF: 11 pages
Proc. SPIE 9121, Multisensor, Multisource Information Fusion: Architectures, Algorithms, and Applications 2014, 91210K (22 May 2014); doi: 10.1117/12.2049947
Show Author Affiliations
Herbert E. M. Viggh, MIT Lincoln Lab. (United States)
Peter L. Cho, MIT Lincoln Lab. (United States)
Nicholas Armstrong-Crews, MIT Lincoln Lab. (United States)
Myra Nam, MIT Lincoln Lab. (United States)
Danelle C. Shah, MIT Lincoln Lab. (United States)
Geoffrey E. Brown, MIT Lincoln Lab. (United States)


Published in SPIE Proceedings Vol. 9121:
Multisensor, Multisource Information Fusion: Architectures, Algorithms, and Applications 2014
Jerome J. Braun, Editor(s)

© SPIE. Terms of Use
Back to Top