AutoBD: Automated Bi-Level Description for Scalable Fine-Grained Visual Categorization.

Abstract

Compared with traditional image classification, fine-grained visual categorization is a more challenging task, because it targets to classify objects belonging to the same species, <i>e.g.</i>, classify hundreds of birds or cars. In the past several years, researchers have made many achievements on this topic. However, most of them are heavily dependent on the artificial annotations, <i>e.g.,</i> bounding boxes, part annotations, <i>and so on</i>. The requirement of artificial annotations largely hinders the scalability and application. Motivated to release such dependence, this paper proposes a robust and discriminative visual description named Automated Bi-level Description (AutoBD). "Bi-level" denotes two complementary part-level and object-level visual descriptions, respectively. AutoBD is "automated," because it only requires the image-level labels of training images and does not need any annotations for testing images. Compared with the part annotations labeled by the human, the image-level labels can be easily acquired, which thus makes AutoBD suitable for large-scale visual categorization. Specifically, the part-level description is extracted by identifying the local region saliently representing the visual distinctiveness. The object-level description is extracted from object bounding boxes generated with a co-localization algorithm. Although only using the image-level labels, AutoBD outperforms the recent studies on two public benchmark, <i>i.e.</i>, classification accuracy achieves 81.6% on CUB-200-2011 and 88.9% on Car-196, respectively. On the large-scale Birdsnap data set, AutoBD achieves the accuracy of 68%, which is currently the best performance to the best of our knowledge.Compared with traditional image classification, fine-grained visual categorization is a more challenging task, because it targets to classify objects belonging to the same species, <i>e.g.</i>, classify hundreds of birds or cars. In the past several years, researchers have made many achievements on this topic. However, most of them are heavily dependent on the artificial annotations, <i>e.g.,</i> bounding boxes, part annotations, <i>and so on</i>. The requirement of artificial annotations largely hinders the scalability and application. Motivated to release such dependence, this paper proposes a robust and discriminative visual description named Automated Bi-level Description (AutoBD). "Bi-level" denotes two complementary part-level and object-level visual descriptions, respectively. AutoBD is "automated," because it only requires the image-level labels of training images and does not need any annotations for testing images. Compared with the part annotations labeled by the human, the image-level labels can be easily acquired, which thus makes AutoBD suitable for large-scale visual categorization. Specifically, the part-level description is extracted by identifying the local region saliently representing the visual distinctiveness. The object-level description is extracted from object bounding boxes generated with a co-localization algorithm. Although only using the image-level labels, AutoBD outperforms the recent studies on two public benchmark, <i>i.e.</i>, classification accuracy achieves 81.6% on CUB-200-2011 and 88.9% on Car-196, respectively. On the large-scale Birdsnap data set, AutoBD achieves the accuracy of 68%, which is currently the best performance to the best of our knowledge.