Share Email Print
cover

Proceedings Paper

The CPHD and R-RANSAC trackers applied to the VIVID dataset
Author(s): Ramona Georgescu; Peter Niedfeldt; Shuo Zhang; Amit Surana; Alberto Speranzon; Ozgur Erdinc
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

In this work, two multitarget trackers - the Cardinalized Probability Hypothesis Density (CPHD) filter and the Recursive Random Sample Consensus (R-RANSAC) algorithm - were applied to three scenarios of the Video Verification of IDentity (VIVID) dataset provided by DARPA. The dataset consists of real video data of multiple cars observed from an unmanned aerial vehicle (UAV) and includes challenging situations such as dense traffic and occlusions. The same detector output was given to each tracker and the same metrics of performance were computed in order to ensure fair comparison of the two tracking approaches. The results show the CPHD did better overall, which was to be expected given that it is the more mature approach.

Paper Details

Date Published: 13 June 2014
PDF: 10 pages
Proc. SPIE 9092, Signal and Data Processing of Small Targets 2014, 90920E (13 June 2014); doi: 10.1117/12.2068075
Show Author Affiliations
Ramona Georgescu, United Technologies Research Ctr. (United States)
Peter Niedfeldt, Brigham Young Univ. (United States)
Shuo Zhang, United Technologies Research Ctr. (United States)
Amit Surana, United Technologies Research Ctr. (United States)
Alberto Speranzon, United Technologies Research Ctr. (United States)
Ozgur Erdinc, United Technologies Research Ctr. (United States)


Published in SPIE Proceedings Vol. 9092:
Signal and Data Processing of Small Targets 2014
Oliver E. Drummond, Editor(s)

© SPIE. Terms of Use
Back to Top