PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Learning Visual Dictionary Using Structured Sparse Principal Component Analysis
Ashish Gupta and Richard Bowden
In: ICIP2012, Florida, USA(2012).


Popular visual category recognition techniques utilize the `Bag-of-Features' method, which represents an image as linear combination of the elements of a learned dictionary. The coefficients of the combination for an image constitute a feature vector which is typically high dimensional and sparse. The high dimensionality and the several uninformative elements in the dictionary together deteriorate classification performance. PCA is used to ameliorate the issues of dimensionality and very recently Sparse PCA (SPCA) has been utilized to compute a comparatively more informative dictionary. In our approach, we adapt Structure Sparse PCA (SSPCA), which leverages encoded higher-order information to build a better dictionary. We present a comprehensive empirical evaluation of our approach against SPCA on several popular datasets and different dictionary sizes.

EPrint Type:Conference or Workshop Item (Other)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Machine Vision
ID Code:9145
Deposited By:Ashish Gupta
Deposited On:21 February 2012