Paper Info
Title
LEAR and XRCE's Participation to Visual Concept Detection Task - ImageCLEF 2010
Abstract
In this paper we present the common effort of Lear and XRCE for the ImageCLEF Visual Concept Detection and Annotation Task. We first sought to combine our individual state-of-the-art approaches: the Fisher vector image representation, with the TagProp method for image auto-annotation. Our second motivation was to investigate the annotation performance by using extra informa- tion in the form of provided Flickr-tags. The results show that using the Flickr-tags in combination with visual features im- proves the results of any method using only visual features. Our winning system, an early-fusion linear-SVM classifier, trained on visual and Flickr-tags features, obtains 45.5% in mean Average Precision (mAP), almost a 5% absolute improve- ment compared to the best visual-only system. Our best visual-only system ob- tains 39.0% mAP, and is close to the best visual-only system. It is a late-fusion linear-SVM classifier, trained on two types of visual features (SIFT and colour). The performance of TagProp is close to our SVM classifiers. The methods presented in this paper, are all scalable to large datasets and/or many concepts. This is due to the fast FK framework for image representation, and due to the classifiers. The linear SVM classifier has proven to scale well for large datasets. The k-NN approach of TagProp, is interesting in this respect since it requires only 2 parameters per concept.
Year
Venue
Keywords
2010
CLEF (Notebook Papers/LABs/Workshops)
fisher vectors,linear svm,multi-modal,tagprop,auto annotation,image classification,machine vision,multi modal,mean average precision
Field
DocType
Citations 
Scale-invariant feature transform,Annotation,Fisher vector,Pattern recognition,Computer science,Support vector machine,Artificial intelligence,Contextual image classification,Classifier (linguistics),Machine learning,Scalability,Linear svm
Conference
11
PageRank 
References 
Authors
0.76
14
5
Name
Order
Citations
PageRank
Thomas Mensink12354116.33
Gabriela Csurka297285.08
Florent Perronnin35448291.48
Jorge Sánchez4282.96
J. J. Verbeek53944181.44