Next Article in Journal
An Improved Belief Entropy to Measure Uncertainty of Basic Probability Assignments Based on Deng Entropy and Belief Interval
Previous Article in Journal
Universal Sample Size Invariant Measures for Uncertainty Quantification in Density Estimation
Open AccessArticle

Distribution Structure Learning Loss (DSLL) Based on Deep Metric Learning for Image Retrieval

1
College of Computer Science and Technology, Jilin University, Changchun 130012, China
2
Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun 130012, China
3
Editorial Department of Journal (Engineering and Technology Edition), Jilin University, Changchun 130012, China
4
School of Computer Science and Engineering, Changchun University of Technology, Changchun 130012, China
*
Author to whom correspondence should be addressed.
Entropy 2019, 21(11), 1121; https://doi.org/10.3390/e21111121
Received: 15 October 2019 / Revised: 9 November 2019 / Accepted: 11 November 2019 / Published: 15 November 2019
The massive number of images demands highly efficient image retrieval tools. Deep distance metric learning (DDML) is proposed to learn image similarity metrics in an end-to-end manner based on the convolution neural network, which has achieved encouraging results. The loss function is crucial in DDML frameworks. However, we found limitations to this model. When learning the similarity of positive and negative examples, the current methods aim to pull positive pairs as close as possible and separate negative pairs into equal distances in the embedding space. Consequently, the data distribution might be omitted. In this work, we focus on the distribution structure learning loss (DSLL) algorithm that aims to preserve the geometric information of images. To achieve this, we firstly propose a metric distance learning for highly matching figures to preserve the similarity structure inside it. Second, we introduce an entropy weight-based structural distribution to set the weight of the representative negative samples. Third, we incorporate their weights into the process of learning to rank. So, the negative samples can preserve the consistency of their structural distribution. Generally, we display comprehensive experimental results drawing on three popular landmark building datasets and demonstrate that our method achieves state-of-the-art performance. View Full-Text
Keywords: deep metric learning; entropy weight; fine-tune network; image retrieval; structural preservation; structural ranking consistency deep metric learning; entropy weight; fine-tune network; image retrieval; structural preservation; structural ranking consistency
Show Figures

Figure 1

MDPI and ACS Style

Fan, L.; Zhao, H.; Zhao, H.; Liu, P.; Hu, H. Distribution Structure Learning Loss (DSLL) Based on Deep Metric Learning for Image Retrieval. Entropy 2019, 21, 1121.

Show more citation formats Show less citations formats
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop