Technical Program

Paper Detail

Paper:	IMDSP-P4.10
Session:	Image/Video Indexing and Retrieval
Time:	Tuesday, May 16, 16:30 - 18:30
Presentation:	Poster
Topic:	Image and Multidimensional Signal Processing: Feature Extraction and Analysis
Title:	LDA versus MMD Approximation on Mislabeled Images for Keyword Dependant Selection of Visual Features and their Heterogeneity
Authors:	Sabrina Tollari, Hervé Glotin, UMR CNRS 6168 LSIS System and Information Sciences lab, France
Abstract:	We propose first to generate new visual features based on entropy measure (heterogeneity), and then we address the question of feature selection in the context of mislabeled images for automatic image classification. We compare two methods of word dependant feature selection on mislabeled images: Approximation of Linear Discriminant Analysis (ALDA) and Approximation of Maximum Marginal Diversity (AMMD). A Hierarchical Ascendant Classification (HAC) is trained and tested using full or reduced visual space. Experiments are conducted on 10K Corel images with 52 keywords, 40 visual features (U) and 40 new heterogeneity features (H). Compared to HAC on all U features, we measure a classification gain of 56% and in the same time a reduction of 92% of the number of features using a simple late fusion of U and H.