Joint Sparse Representation and Robust Feature-Level Fusion for Multi-Cue Visual Tracking
Authors:
Open Access
- 23 September 2015
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 24 (12) , 5826-5841
Abstract
Visual tracking using multiple features has been proved as a robust approach because features could complement each other. Since different types of variations such as illumination, occlusion, and pose may occur in a video sequence, especially long sequence videos, how to properly select and fuse appropriate features has become one of the key problems in this approach. To address this issue, this paper proposes a new joint sparse representation model for robust feature-level fusion. The proposed method dynamically removes unreliable features to be fused for tracking by using the advantages of sparse representation. In order to capture the non-linear similarity of features, we extend the proposed method into a general kernelized framework, which is able to perform feature fusion on various kernel spaces. As a result, robust tracking performance is obtained. Both the qualitative and quantitative experimental results on publicly available videos show that the proposed method outperforms both sparse representation-based and fusion based-trackers.Keywords
Funding Information
- General Research Fund through the Research Grants Council, Hong Kong (HKBU 212313)
This publication has 68 references indexed in Scilit:
- Multiple Kernel Learning for Sparse Representation-Based ClassificationIEEE Transactions on Image Processing, 2014
- Log-Euclidean Kernels for Sparse Representation and Dictionary Learning2013
- Tracking via Robust Multi-task Multi-view Joint Sparse Representation2013
- Sparse Representation With KernelsIEEE Transactions on Image Processing, 2012
- Visual Classification With Multitask Joint Sparse RepresentationIEEE Transactions on Image Processing, 2012
- Distribution fields for tracking2012
- Robust Visual Tracking Using an Effective Appearance Model Based on Sparse CodingACM Transactions on Intelligent Systems and Technology, 2012
- Tracking-Learning-DetectionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2011
- Hough-based tracking of non-rigid objects2011
- Accelerated Gradient Method for Multi-task Sparse Learning Problem2009