Sketch-based cross-domain image retrieval via heterogeneous network
Sketch-based cross-domain image retrieval via heterogeneous network
复制标题
DOI：
10.1109/vcip.2017.8305153
复制
发表时间：
2017-12
期刊：
影响因子：
--
通讯作者：
H. Zhang;Chuang Zhang;Ming Wu
中科院分区：
文献类型：
--
作者：
H. Zhang;Chuang Zhang;Ming Wu
图像多样性的发展带动了跨域图像检索在多个领域的应用。在本文中，我们提出了一个基于草图和图像的异构双重网络（两个不同的网络）。通过对比损失与三组排序损失相结合的方法，限制从两个网络中提取的特征的相似性，实现端到端的跨域图像检索。我们还研究了绘图顺序对草图检索的影响。与Siamese网络相比，我们避免了边缘提取的预处理，进一步增强了细粒度检索的效果。
The development of image diversity has led multiple fields' application of cross-domain image retrieval. In this paper, we propose a heterogeneous dual network (two different networks) based on sketches and images. End-to-end cross-domain image retrieval is realized by limiting the similarity of features extracted from the two networks through the method of combining the contrastive loss with the triplet ranking loss. We also study how the order of drawings affects the sketch retrieval. Compared to the Siamese network, we avoid edge extraction for preprocessing, and the effect of fine-grained retrieval is further enhanced.