E.: Large-scale category structure aware image categorization

Most previous research on image categorization has focused on medium-scale data sets, while large-scale image categorization with millions of images from thousands of categories remains a challenge. With the emergence of structured large-scale dataset such as the ImageNet, rich information about the conceptual relationships between images, such as a tree hierarchy among various image cat-egories, become available. As human cognition of complex visual world benefits from underlying semantic relationships between object classes, we believe a ma-chine learning system can and should leverage such information as well for better performance. In this paper, we employ such semantic relatedness among image categories for large-scale image categorization. Specifically, a category hierarchy is utilized to properly define loss function and select common set of features for related categories. An efficient optimization method based on proximal approxi-mation and accelerated parallel gradient method is introduced. Experimental re-sults on a subset of ImageNet containing 1.2million images from 1000 categories demonstrate the effectiveness and promise of our proposed approach.