点击次数：

论文类型：会议论文

发表时间：2019-01-01

收录刊物：EI、CPCI-S

卷号：2019-July

页面范围：634-639

关键字：Fine-grained image classification; Self-regressive localization; Discriminative prior

摘要：Currently, most top-performing Weakly supervised Fine-grained Image Classification (WFGIC) schemes tend to pick out discriminative patches. However, those patches usually contain much noise information, which influences the accuracy of the classification. Besides, they rely on a large amount of candidate patches to discover the discriminative ones, thus leading to high computational cost. To address these problemes, we propose a novel end-to-end Self-regressive Localization with Discriminative Prior Network (SDN) model, which learns to explore more accurate size of discriminative patches and enables to classify images in real time. Specifically, we design a multi-task discriminative learning network, a self-regressive localization sub-network and a discriminative prior sub-network with the guided loss as well as the consistent loss to simultaneously learn self-regressive coefficients and discriminative prior maps. The self-regressive coefficients can decrease noise information in discriminative patches and the discriminative prior maps through learning discriminative probability values filter thousands of canditate patches to single figure. Extensive experiments demonstrate that the proposed SDN model achieves state-of-the-art both in accuracy and efficiency.