基于注意力金字塔与监督哈希的细粒度图像检索

Fine-grained Image Retrieval Based on Supervised Hashing with Attention Pyramid

大规模细粒度图像检索是一项极具挑战性的任务。由于图像间具有类间距离小、类内距离大的特点,传统的深度神经网络学习到的图像特征存在高度冗余,导致检索速度慢、存储成本高昂。为解决该问题,提出了一种基于注意力金字塔与监督哈希的深度神经网络模型。在特征提取网络中,针对细粒度图像的特点,采用了双通路金字塔结构,并设计了自上而下的特征通路及自下而上的注意力通路,借此更好地融合高层与低层特征。在分类网络中,为压缩存储空间、提高检索效率,在深度哈希的基础上使用tanh(x)代替sign(x)作为激活函数,使学习到的哈希函数更容易达到平稳分布;同时结合量化损失与分类损失,使生成的哈希码更好地与原始输入图像的特征匹配。在FGVC-Aircraft及Stanford Cars两个标准细粒度数据集上的准确率分别达到82.3%、83.3%,均优于其他对比算法,证明了算法的有效性。

Large-scale fine-grained image retrieval is a challenging task. Due to the small inter-class variations and the large intra-class variations among images, features learned by traditional CNNs is highly redundant, which results in slow query speed and expensive storage cost. To address this problem, we propose a novel convolutional neural network which combines attention pyramid and supervised hashing. Specifically, in order to extract finer features, we introduce a dual pathway hierarchy structure in the feature extraction network with a top-down feature pathway and a bottom-up attention pathway, which is utilized to combine high-level semantic information and low-level detailed feature representations. Furthermore, to reduce storage cost and increase query speed, we improve deep hashing by using tanh(x) instead of sign(x) as the activation function to make sure that the learned hash function achieves stable distribution. At the same time, we adopt both quantization loss and classification loss to map the binary codes to the origin images better. The experimental results demonstrate that the proposed algorithm is superior to other comparison algorithms, for it achieves 82.3% and 83.3% accuracy on the FGVC-Aircraft and the Stanford Cars test set, which proves the effectiveness of the algorithm.