Abstract : A new image representation based on distribution of local invariant features to be used in a discriminative approach to image categorization is presented. The representation which is called Probability Signature (PS) is combined with character of two distribution models Probability Density Function and standard signatures. The PS representation retains high discriminative power of PDF model, and is suited for measuring dissimilarity of images with Earth Mover's Distance (EMD), which allows for partial matches of compared distributions. It is evaluated on whole-image classification tasks from the scene and category image datasets. The comparative experiments show that the proposed algorithm has inspiring performance.
Zhongzhi Shi; Sunil Vadera; Agnar Aamodt; David Leake. 6th IFIP TC 12 International Conference on Intelligent Information Processing (IIP), Oct 2010, Manchester, United Kingdom. Springer, IFIP Advances in Information and Communication Technology, AICT-340, pp.308-315, 2010, Intelligent Information Processing V. 〈10.1007/978-3-642-16327-2_37〉
https://hal.inria.fr/hal-01060367
Contributeur : Hal Ifip
<>
Soumis le : jeudi 4 septembre 2014 - 12:56:09
Dernière modification le : vendredi 11 août 2017 - 14:50:58
Document(s) archivé(s) le : vendredi 5 décembre 2014 - 10:11:34
Lijun Guo, Jieyu Zhao, Rong Zhang. A Novel Distribution of Local Invariant Features for Classification of Scene and Object Categories. Zhongzhi Shi; Sunil Vadera; Agnar Aamodt; David Leake. 6th IFIP TC 12 International Conference on Intelligent Information Processing (IIP), Oct 2010, Manchester, United Kingdom. Springer, IFIP Advances in Information and Communication Technology, AICT-340, pp.308-315, 2010, Intelligent Information Processing V. 〈10.1007/978-3-642-16327-2_37〉. 〈hal-01060367〉
