Clothing fashion style recommendation system

Permanent URL: http://hdl.handle.net/2047/D20193616
Title:
Clothing fashion style recommendation system
Creator:
Dai, Wei (Author)
Contributor:
Fu, Yun (Advisor)
Liu, Yongmin (Committee member)
Kong, Yu (Committee member)
Language:
English
Publisher:
Boston, Massachusetts : Northeastern University, 2015
Date Accepted:
May 2015
Date Awarded:
August 2015
Type of resource:
Text
Genre:
Masters theses
Format:
electronic
Digital origin:
born digital
Abstract/Description:
This thesis proposes a clothing recommendation system that can recommend clothing images based on the fashion style of the provided clothing images. In this work, we focus on the images of upper body clothing and with human model in the images.

In the first part, we present a clothing dataset collected from Internet containing 27,375 men's and women's clothing images of 11 clothing categories. We develop a recommendation system that can differentiate fashion categories of query images. We propose a framework that divides the system into three decoupled and autonomous components in order to provide a highly flexible and extensible system. Then we describe an implementation of this framework on a Linux server. To demonstrate this clothing recommendation system we also develop two user interfaces, including a Web Application and an iOS App. Lastly, we discuss the approaches to secure the system and user privacy.

We set up a Demo of this clothing recommendation system running on iPhone, which can achieve promising results within 5 seconds.
Subjects and keywords:
clothing
clothing recommendation system
web application
Optical pattern recognition
Content-based image retrieval
Image processing -- Digital techniques
Database design
Decoupling (Mathematics)
Application software -- Development
Mobile apps -- Design
Fashion -- Databases
Clothing and dress -- Databases
DOI:
https://doi.org/10.17760/D20193616
Permanent Link:
http://hdl.handle.net/2047/D20193616
Use and reproduction:
In Copyright: This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the right-holder(s). (http://rightsstatements.org/vocab/InC/1.0/)
Copyright restrictions may apply.

Downloads