Discriminative cellets discovery for fine-grained image categories retrieval

Publisher:
ACM
Publication Type:
Conference Proceeding
Citation:
ICMR 2014 - Proceedings of the ACM International Conference on Multimedia Retrieval 2014, 2014, pp. 57 - 64
Issue Date:
2014-01-01
Full metadata record
Files in This Item:
Filename Description Size
p57-Zhang.pdfPublished version5.63 MB
Adobe PDF
Fine-grained image categories recognition is a challenging task aiming at distinguishing objects belonging to the same basic-level category, such as leaf or mushroom. It is a useful technique that can be applied for species recognition, face verification, and etc. Most of the existing methods have difficulties to automatically detect discriminative object components. In this paper, we propose a new fine- grained image categorization model that can be deemed as an improved version spatial pyramid matching (SPM). In- stead of the conventional SPM that enumeratively conducts cell-to-cell matching between images, the proposed model combines multiple cells into cellets that are highly responsive to object fine-grained categories. In particular, we describe object components by cellets that connect spatially adjacent cells from the same pyramid level. Straightforwardly, image categorization can be casted as the matching between cellets extracted from pairwise images. Toward an effective matching process, a hierarchical sparse coding algorithm is derived that represents each cellet by a linear combination of the basis cellets. Further, a linear discriminant analy- sis (LDA)-like scheme is employed to select the cellets with high discrimination. On the basis of the feature vector built from the selected cellets, fine-grained image categorization is conducted by training a linear SVM. Experimental results on the Caltech-UCSD birds, the Leeds butterflies, and the COSMIC insects data sets demonstrate our model out- performs the state-of-the-art. Besides, the visualized cellets show discriminative object parts are localized accurately. Copyright 2014 ACM.
Please use this identifier to cite or link to this item: