Home // ICIW 2015, The Tenth International Conference on Internet and Web Applications and Services // View article
Authors:
Nodoka Yamamoto
Shuichi Kurabayashi
Keywords: Graphic design; Color scheme; Image processing; Recommendation system
Abstract:
We propose a novel Web-based graphic design system that utilizes data-driven techniques to share and reuse the practitioners’ knowledge of graphic design. A unique feature of this system is a user interaction model that utilizes an image being edited as a query for the knowledge base in order to recommend colors suited to it. The system stores images using a highly abstract graph structure, which we refer to as an Abstract Appearance Graph (AAG). The content being drawn at any given time is also modeled as an AAG. We have thus implemented a vector graphics design environment as an interactive graph database retrieval process to reuse existing designs. Our current prototype supports the extraction and reuse of the appearance properties of AAGs: the gradient of colors, their size, and positional relations. We implemented a cloud-based prototype system to create and share AAGs by using modern HTML5 technology in order to show the feasibility of our approach. We also performed experimental studies with impressional (kansei) graphic design knowledge. The experimental result shows that our system effectively supports graphic design by bringing to bear existing knowledge on new designs and rendering graphic design more flexible.
Pages: 7 to 15
Copyright: Copyright (c) IARIA, 2015
Publication date: June 21, 2015
Published in: conference
ISSN: 2308-3972
ISBN: 978-1-61208-412-1
Location: Brussels, Belgium
Dates: from June 21, 2015 to June 26, 2015