Efficient Supervised Dimensionality Reduction for Image Categorization

Rachid Benmokhtar (1) , Jonathan Delhumeau (1) , Philippe-Henri Gosselin (2)

1 TEXMEX - Multimedia content-based indexing
2 ETIS - UMR 8051 - Equipes Traitement de l'Information et Systèmes

This paper addresses the problem of large scale image repre- sentation for object recognition and classification. Our work deals with the problem of optimizing the classification accu- racy and the dimensionality of the image representation. We propose to iteratively select sets of projections from an ex- ternal dataset, using Bagging and feature selection thanks to SVM normals. Features are selected using weights of SVM normals in orthogonalized sets of projections. The Bagging strategy is employed to improve the results and provide more stable selection. The overall algorithm linearly scales with the size of features, and thus is able to process the large state- of-the-art image representation. Given Spatial Fisher Vectors as input, our method consistently improves the classification accuracy for smaller vector dimensionality, as demonstrated by our results on the popular and challenging PASCAL VOC 2007 benchmark.

Domaines: Machine Learning [stat.ML]; Vision par ordinateur et reconnaissance de formes [cs.CV]

Rachid Benmokhtar, Jonathan Delhumeau, Philippe-Henri Gosselin. Efficient Supervised Dimensionality Reduction for Image Categorization. IEEE International Conference on Acoustics, Speech, and Signal Processing, May 2013, Vancouver, Canada. pp.2425-2428. ⟨hal-00807483⟩