Combined feature evaluation for adaptive visual object tracking

► A combined feature set for object tracking. ► A novel feature evaluation approach considering temporal consistency. ► A new application of traditional tracking algorithms to model feature confidence. Existing visual tracking methods are challenged by object and background appearance variations, wh...

Full description

Saved in:
Bibliographic Details
Published in:Computer vision and image understanding 2011, Vol.115 (1), p.69-80
Main Authors: Han, Zhenjun, Ye, Qixiang, Jiao, Jianbin
Format: Article
Language:English
Subjects:
Applied sciences
Artificial intelligence
C (programming language)
Color
Color histogram
Computer science
control theory
systems
Consistency
Exact sciences and technology
Gradient orientation histogram
Histograms
Kalman filter
Object tracking
Particle filter
Pattern recognition. Digital image processing. Computational geometry
Representations
Tracking
Visual
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:► A combined feature set for object tracking. ► A novel feature evaluation approach considering temporal consistency. ► A new application of traditional tracking algorithms to model feature confidence. Existing visual tracking methods are challenged by object and background appearance variations, which often occur in a long duration tracking. In this paper, we propose a combined feature evaluation approach in filter frameworks for adaptive object tracking. First, a feature set is constructed by combining color histogram (HC) and gradient orientation histogram (HOG), which gives a representation of both color and contour. Then, to adapt to the appearance changes of the object and its background, these features are assigned with different confidences adaptively to make the features with higher discriminative ability play more important roles in the instantaneous tracking. To keep the temporal consistency, the feature confidences are evaluated based on Kalman and Particle filters. Experiments and comparisons demonstrate that object tracking with evaluated features have good performance even when objects go across complex backgrounds.
ISSN:1077-3142
1090-235X
DOI:10.1016/j.cviu.2010.09.004