Visual prior from generic real-world images can be learned and transferred for representing objects in a scene. Motivated by this, we propose an algorithm that transfers visual prior learned offline for online object tracking. From a collection of real-world images, we learn an overcomplete dictionary to represent visual prior. The prior knowledge of objects is generic, and the training image set does not necessarily contain any observation of the target object. During the tracking process, the learned visual prior is transferred to construct an object representation by sparse coding and multiscale max pooling. With this representation, a linear classifier is learned online to distinguish the target from the background and to account for the target and background appearance variations over time. Tracking is then carried out within a Bayesian inference framework, in which the learned classifier is used to construct the observation model and a particle filter is used to estimate the tracking result sequentially. Experiments on a variety of challenging sequences with comparisons to several state-of-the-art methods demonstrate that more robust object tracking can be achieved by transferring visual prior.
Bibliographical noteFunding Information:
Manuscript received July 08, 2011; revised October 21, 2011 and February 14, 2012; accepted February 24, 2012. Date of publication April 05, 2012; date of current version June 13, 2012. This work was supported in part by a Google Faculty Award, by NSF CAREER under Grant 1149783, and by NSF IIS under Grant 1152576. The work of Q. Wang and F. Chen was supported in part by the National Natural Science Foundation of China under Grant 61071131 and in part by the Beijing Natural Science Foundation under Grant 4122040. The work of J. Yang and M.-H. Yang was supported in part by the National Science Foundation (NSF) CAREER under Grant 1149783 and by the NSF Division of Information and Intelligent Systems (IIS) under Grant 1152576. The work of W. Xu was supported in part by the the National Key Basic Research and Development Program of China under Grant 2009CB320602. The associate editor coordinating the review of this manuscript and approving it for publication was Prof. Kenneth K. M. Lam.
All Science Journal Classification (ASJC) codes
- Computer Graphics and Computer-Aided Design