Visual Tracking Using 3D Data and Region-Based Active Contours

Abstract

Our research was concerned with the theory and practice of visual information in a feedback loop, the underlying problem of controlled active vision. Controlled active vision requires the integration of techniques from control theory, signal processing, and computer vision. Visual tracking provides an important example of the need for controlled active vision. Tracking in the presence of a disturbance is a classical control issue, but because of the highly uncertain nature of the disturbance, this type of problem is very difficult. Visual tracking differs from standard tracking problems since the feedback signal is measured using imaging sensors. In particular, it has to be extracted via computer vision and image processing algorithms and interpreted by a reasoning algorithm before being used in the control loop. The development of robust, reliable visual tracking algorithms will be one of the main challenges in our proposed research.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 28, 2016
Accession Number
AD1053584

Entities

People

  • Allen Tannenbaum

Organizations

  • Georgia Tech Research Corporation

Tags

Communities of Interest

  • Materials and Manufacturing Processes
  • Sensors

DTIC Thesaurus Topics

  • Algorithms
  • Computational Complexity
  • Computer Vision
  • Computers
  • Control Systems
  • Control Theory
  • Data Displays
  • Differential Equations
  • Filters
  • Filtration
  • Image Processing
  • Mathematical Filters
  • Partial Differential Equations
  • Sequential Monte Carlo Methods
  • Signal Processing
  • Standards
  • Statistical Estimation

Fields of Study

  • Computer science

Readers

  • Artificial Intelligence
  • Computer Vision.
  • Control Systems Engineering.

Technology Areas

  • AI & ML
  • AI & ML - Machine Learning Algorithms