Deep Semantic Ranking Based Hashing for
Multi-Label Image Retrieval
 |
| The proposed deep semantic ranking based hashing. |
People
Fang Zhao
Yongzhen Huang
Liang Wang
Tieniu Tan
Overview
With the rapid growth of web images, hashing has received increasing interests in large scale image retrieval. Research efforts have been devoted to learning compact binary codes that preserve semantic similarity based on labels. However, most of these hashing methods are designed to handle simple binary similarity. The complex multilevel semantic structure of images associated with multiple labels have not yet been well explored. Here we propose a deep semantic ranking based method for learning hash functions that preserve multilevel semantic similarity between multilabel images. In our approach, deep convolutional neural network is incorporated into hash functions to jointly learn feature representations and mappings from them to hash codes, which avoids the limitation of semantic representation power of hand-crafted features. Meanwhile, a ranking list that encodes the multilevel similarity information is employed to guide the learning of such deep hash functions. An effective scheme based on surrogate loss is used to solve the intractable optimization problem of nonsmooth and multivariate ranking measures involved in the learning procedure. Experimental results show the superiority of our proposed approach over several state-of-theart hashing methods in term of ranking evaluation metrics when tested on multi-label image datasets.
Paper
 |
Deep Semantic Ranking Based Hashing for Multi-Label Image Retrieval Fang Zhao, Yongzhen Huang, Liang Wang, Tieniu Tan Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition (CVPR2015) |
Experimental Results
| Datasets: (a) MIRFLICKR-25K; (b) NUS-WIDE | |
 |
 |
| Comparison of ranking performance of our DSRH and other hashing methods based on hand-crafted features on two datasets. | Comparison of ranking performance of our DSRH and other hashing methods based on activation features of pre-trained CNN on two datasets. |
Acknowledgments
This work was supported by the National Basic Research Program of China (2012CB316300), National Natural Science Foundation of China (61175003, 61135002, 61420106015, U1435221), and CCF-Tencent Open Fund.