[NIPS 2011] Matrix Completion for Multi-label Image Classification英文文献资料.pdf

Matrix Completion for Multi-label Image Classification †,‡ ‡ † † Ricardo S. Cabral Fernando De la Torre João P. Costeira , Alexandre Bernardino ‡Carnegie Mellon University, †ISR - Instituto Superior Técnico, Pittsburgh, PA Lisboa, Portugal rscabral@cmu.edu, ftorre@cs.cmu.edu, {jpc,alex}@isr.ist.utl.pt Abstract Recently, image categorization has been an active research topic due to the urgent need to retrieve and browse digital images via semantic keywords. This paper for- mulates image categorization as a multi-label classification problem using recent advances in matrix completion. Under this setting, classification of testing data is posed as a problem of completing unknown label entries on a data matrix that concatenates training and testing features with training labels. We propose two convex algorithms for matrix completion based on a Rank Minimization criterion specifically tailored to visual data, and prove its convergence properties. A major advantage of our approach w.r.t. standard discriminative classification methods for image categorization is its robustness to outliers, background noise and par- tial occlusions both in the feature and label space. Experimental validation on several datasets shows how our method outperforms state-of-the-art algorithms, while effectively capturing semantic concepts of classes. 1 Introduction With the ever-growing amount of digital image data in multimedia databases, there is a great need for algorithms that can provide