Call For Paper Volume:4 Issue:6 Jun'2017

Search based face annotation Approach for Mining Weakly labeled Web facial Images: A Review

Publication Date : 12/04/2015



Author(s) :

Dipak D. Tayade , Dipak Pardhi.


Volume/Issue :
Volume 2
,
Issue 4
(04 - 2015)



Abstract :

This paper investigates search based face annotation (SFBA) framework with the help of mining weakly labeled web facial images which are freely available on World Wide Web (WWW). The drawback as a problem in SBFA is how to perform effectively annotation by considering ordered list of most similar facial images which are weakly labeled that are often noisy and incomplete. To overcome this problem we present the most effective unsupervised label refinement (ULR) method for refining the labels of web facial images using machine learning techniques. We consider the learning problem as a convex optimization and develop effective optimization algorithms to solve the large-scale learning task efficiently. For the fast processing, we also propose a clustering-based approximation algorithm which can develop the scalability. Keywords- Face annotation, content-based image retrieval, machine learning, label refinement, web facial images, weak label.


No. of Downloads :

3


Indexing

Web Design MymensinghPremium WordPress ThemesWeb Development

Search based face annotation Approach for Mining Weakly labeled Web facial Images: A Review

April 10, 2015