Visual Tracking Decomposition
Authors
Abstract
We propose a novel tracking algorithm that can work robustly in a challenging scenario such that several kinds of appearance and motion changes of an object occur at the same time. Our algorithm is based on a visual tracking decomposition scheme for the efficient design of observation and motion models as well as trackers. In our scheme, the observation model is decomposed into multiple basic observation models that are constructed by sparse principal component analysis (SPCA) of a set of feature templates. Each basic observation model covers a specific appearance of the object. The motion model is also represented by the combination of multiple basic motion models, each of which covers a different type of motion. Then the multiple basic trackers are designed by associating the basic observation models and the basic motion models, so that each specific tracker takes charge of a certain change in the object. All basic trackers are then integrated into one compound tracker through an interactive Markov Chain Monte Carlo (IMCMC) framework in which the basic trackers communicate with one another interactively while run in parallel. By exchanging information with others, each tracker further improves its performance, which results in increasing the whole performance of tracking. Experimental results show that our method tracks the object accurately and reliably in realistic videos where the appearance and motion are drastically changing over time.
Paper
CVPR 2010 paper. (pdf, 2.4MB) PPT. (ppt, 5MB)
Citation
1. Junseok Kwon, Kyoung Mu Lee. Visual Tracking Decomposition, IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2010 Bibtex
2. Junseok Kwon, Kyoung Mu Lee. Tracking by Sampling Trackers, IEEE International Conference on Computer Vision (ICCV) 2011 [project page]
Results
video. (wmv, 40.9MB)Code
source code . (zip, 52MB, version 7.0)binary code . (zip, 243MB, version 0.61)
dataset :images. (zip, 290.9MB)
dataset :videos. (zip, 1.6GB)