Abstract
User-provided tags associated with social images are essential information for social image retrieval. Unfortunately, these tags are often imperfect to describe the visual contents, which severely degrades the performance of image retrieval. Tag relevance learning models are proposed to improve the descriptive powers of tags mostly based on the Gaussian noise assumption. However, the intrinsic probability distribution of the noise is unknown and other probability distributions may be much better. Towards this end, this paper investigates the applicable probability distributions of tag noise and proposes a novel Cauchy Matrix Factorization (CMF) method for tag-based image retrieval. The Cauchy probability distribution is robust to all kinds of noise and more suitable to model the tagging noise of social images. Therefore, we utilize Cauchy distribution to model noise under the matrix factorization framework. Besides, other five probability density functions, i.e., Gaussian, Laplacian, Poisson, Student-t and Logistic, are investigated to model noise of social tags. To evaluate the performance of different probability distributions, extensive experiments on two widely-used datasets are conducted and results show the robustness of CMF to noisy tags of social images.
Highlights
Recent years have witnessed the explosive growth of social images associated with user-provided tags, which often makes users difficult to find their desired images
This paper proposes a novel Cauchy Matrix Factorization (CMF) method by modeling the tagging noise using the Cauchy distribution
We deeply investigate the applicable probability distributions based on different probabilistic noise assumptions, i.e., Laplacian, Poisson, Student-t, Logistic and Cauchy distributions, and find the Cauchy assumptions is robust to all kinds of noise
Summary
Recent years have witnessed the explosive growth of social images associated with user-provided tags, which often makes users difficult to find their desired images. Many methods have been proposed for image retagging and tag refinement by removing the noisy tags and complementing the relevant but missing tags [18]–[20], [23], [27], [30], [33], [35], [37]–[39] Most of these methods learn the image-tag relevance by minimizing the prediction error based on the Matrix Factorization (MF) framework. It is necessary and important to propose a new MF method to deal with various types of noise for tag-based image retrieval. Towards this end, this paper proposes a novel Cauchy Matrix Factorization (CMF) method by modeling the tagging noise using the Cauchy distribution. The tradition matrix factorization model is based on the Gaussian noise assumption, which leads to the minimizing optimization problem with the sum-of-squared-errors objective function. Extensive experiments are conducted on two widely-used social image benchmarks for tag-based social image retrieval
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
