Visual Tracking via Online Nonnegative Matrix Factorization
Authors:
- 15 August 2013
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 24 (3) , 374-383
Abstract
In visual tracking, holistic and part-based representations are both popular choices to model target appearance. The former is known for great efficiency and convenience, while the latter for robustness against local appearance or shape variations. Based on nonnegative matrix factorization (NMF), we propose a novel visual tracker that takes advantage of both groups. The idea is to model the target appearance by a nonnegative combination of nonnegative components learned from examples observed in previous frames. To adjust NMF to the tracking context, we include sparsity and smoothness constraints in addition to the nonnegativity one. Furthermore, an online iterative learning algorithm, together with a proof of convergence, is proposed for efficient model updating. Putting these ingredients together with a particle filter framework, the proposed tracker, constrained online nonnegative matrix factorization (CONMF), achieves robustness to challenging appearance variations and nontrivial deformations while running in real time. We evaluate the proposed tracker on various benchmark sequences containing targets undergoing large variations in scale, pose, or illumination. The robustness and efficiency of CONMF is validated in comparison with several state-of-the-art trackers.Keywords
Funding Information
- NSFC (61005027, 61273259, 61272223, 61370036)
- NSF of Jiangsu Province (BK2012045, BK2011825)
- NSF (IIS-1218156)
This publication has 32 references indexed in Scilit:
- A survey of appearance models in visual object trackingACM Transactions on Intelligent Systems and Technology, 2013
- Efficient Minimum Error Bounded Particle Resampling L1 Tracker With Occlusion DetectionIEEE Transactions on Image Processing, 2013
- Incremental subspace learning via non-negative matrix factorizationPattern Recognition, 2009
- Incremental Learning for Robust Visual TrackingInternational Journal of Computer Vision, 2007
- Sparse non-negative matrix factorizations via alternating non-negativity-constrained least squares for microarray data analysisBioinformatics, 2007
- Object trackingACM Computing Surveys, 2006
- The template update problemIEEE Transactions on Pattern Analysis and Machine Intelligence, 2004
- Robust online appearance models for visual trackingIEEE Transactions on Pattern Analysis and Machine Intelligence, 2003
- Kernel-based object trackingIEEE Transactions on Pattern Analysis and Machine Intelligence, 2003
- Efficient region tracking with parametric models of geometry and illuminationIEEE Transactions on Pattern Analysis and Machine Intelligence, 1998