The Robotics Institute
RI | VASC | VIVID | Tracking Evaluation

 VIVID Tracking Evaluation Web Site


Home Datasets Software Submission Results  

 Overview

The VIVID Tracking Evaluation Web Site was developed to promote sharing and evaluation of algorithms that track objects through video sequences. Development of this page was suggested and sponsored by Dr. Thomas Strat from DARPA IXO, under the Video Verification of Identity (VIVID) program. The initial application of interest is tracking ground vehicles from airborne sensor platforms.

The primary function of this web site is to provide:
1) Ground-truthed tracking data sets;
2) An open-source tracking testbed that includes code for several baseline tracking algorithms;
3) A mechanism for uploading tracking results for automated scoring against ground-truth.

Although users are required to create an account to submit results, no account is necessary to download data sets, the tracking testbed, or to view currently ranked algorithm results.

 Publications

We grant permission to use and publish all data and results from this site, provided the following paper is referenced:

  Personnel

Name Title Email Address
Robert T. Collins Assoc. Research Professor
rcollins@cs.cmu.edu
Martial Hebert Professor
hebert@ri.cmu.edu

Hulya Yalcin Postdoc Research Associate Hulya_Yalcin@brown.edu

David Tolliver Graduate Student tolliver@cs.cmu.edu

Marius Leordeanu Graduate Student mleordeanu@cs.cmu.edu

Xuhui Zhou Research Programmer xuhui@cs.cmu.edu
Seng Keat Teh Undergrad steh@cs.cmu.edu

  Related Sites

Associated center: VIVID, VASC

Video Verification of Identity (VIVID) We address the problem of tracking object in video sequences and of recognizing the tracked objects in subsequent sequences.
Vision and Autonomous System Center (VASC)  - The main umbrella group within Robotics Institue, working in the areas of computer vision, autonomous navigation, virtual reality, intelligent manipulation, space robotics, and related fields.