Title :
Discriminative spatial saliency for image classification
Author :
Sharma, Gaurav ; Jurie, Frédéric ; Schmid, Cordelia
Author_Institution :
GREYC, Univ. de Caen, Caen, France
Abstract :
In many visual classification tasks the spatial distribution of discriminative information is (i) non uniform e.g. person `reading´ can be distinguished from `taking a photo´ based on the area around the arms i.e. ignoring the legs and (ii) has intra class variations e.g. different readers may hold the books differently. Motivated by these observations, we propose to learn the discriminative spatial saliency of images while simultaneously learning a max margin classifier for a given visual classification task. Using the saliency maps to weight the corresponding visual features improves the discriminative power of the image representation. We treat the saliency maps as latent variables and allow them to adapt to the image content to maximize the classification score, while regularizing the change in the saliency maps. Our experimental results on three challenging datasets, for (i) human action classification, (ii) fine grained classification and (iii) scene classification, demonstrate the effectiveness and wide applicability of the method.
Keywords :
feature extraction; gesture recognition; image classification; image representation; learning (artificial intelligence); classification score maximization; discriminative information; discriminative spatial saliency learning; fine grained classification; human action classification; image classification; image content adaptation; image representation; intraclass variation; latent variables; max margin classifier; scene classification; spatial distribution; visual classification task; visual feature; Histograms; Humans; Optimization; Support vector machines; Training; Vectors; Visualization;
Conference_Titel :
Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on
Conference_Location :
Providence, RI
Print_ISBN :
978-1-4673-1226-4
Electronic_ISBN :
1063-6919
DOI :
10.1109/CVPR.2012.6248093