BMI | talk9
16371
post-template-default,single,single-post,postid-16371,single-format-standard,ajax_fade,page_not_loaded,,side_area_uncovered_from_content,qode-child-theme-ver-1.0.0,qode-theme-ver-11.0,qode-theme-bridge,wpb-js-composer js-comp-ver-5.1.1,vc_responsive
 

talk9

talk9

Zhou Wei

Ph.D. , Chongqing University

August 14, 2017

Abstract

Information overload problem exists in our daily life, so it is in fashion e-commerce area. There are some unique attributes in fashion recommendation. How to mine useful information in textual description and fashion product images is the key to get better recommendation. In this talk, a fashion products recommendation method is proposed, including similar product and mix-match product recommendation. First, a recategorization method of fashion products are proposed based on deep learning techniques, products are recategorized into the defined categories; secondly, features are extracted from these images, including color palette of each image, and texture features of images; finally, a content based recommendation method is proposed. In this talk, Dr. Zhou will introduce his current work, some interesting future work will also be introduced about fashion recommendation.

More…

 

No Comments

Sorry, the comment form is closed at this time.