Co-training Framework of Generative and Disciminative Trackers with Partial Occlusion Handling

Abstract

Partial occlusion is a challenging problem in object tracking. In online visual tracking, it is the critical factor causing drift. To address this problem, we propose a novel approach using a co-training framework of generative and discriminative trackers. Our approach is able to detect the occluding region and continuously update both the generative and discriminative models using the information from the non-occluded part. The generative model encodes all of the appearance variations using a low dimension subspace which helps provide a strong reacquisition ability. Meanwhile the discriminative classifier, an online support vector machine, focuses on separating the object from the background using a Histograms of Oriented Gradients (HOG) feature set. For each search window, an occlusion likelihood map is generated by the two trackers through a codecision process. If there is disagreement between these two trackers, the movement vote of KLT local features is used as a referee. Precise occlusion segmentation is performed using MeanShift. Finally, each tracker recovers the occluded part and updates its own model using the new nonoccluded information. Experimental results on challenging sequences with different types of objects are presented. We also compare with other state-of-the-art methods to demonstrate the superiority and robustness of our tracking framework.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2011
Accession Number
ADA536863

Entities

People

  • Gerard Medioni
  • Thang Ba Dinh

Organizations

  • University of Southern California

Tags

Communities of Interest

  • Air Platforms
  • Energy and Power Technologies
  • Human Systems
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Algorithms
  • Artificial Intelligence Software
  • Bayesian Networks
  • Computer Programs
  • Computer Vision
  • Detection
  • Detectors
  • Distance Learning
  • Education
  • Generative Models
  • Histograms
  • Machine Learning
  • Models
  • Sequences
  • Supervised Machine Learning
  • Training
  • Vascular System Injuries

Fields of Study

  • Computer science

Readers

  • Adaptive Control and Estimation with Uncertainty in Dynamic Systems.
  • Agent-Based Social Robotics and Mobile-Assisted Learning in Virtual Environments.
  • Computer Vision.

Technology Areas

  • AI & ML