Mobile computer vision system for document image rendering, manipulation, and management on collaborative e-writing devices

点击翻译按钮获取中文摘要

The next generation of business productivity tools focuses on supporting creative processes underlying innovation. Recent studies show that while creativity is perceived as a key asset in business performance, many companies do not consider themselves as being creative. There is a clear need for enabling the manifestation of creative potential with novel business productivity tools. Our partner company, QuirkLogic Inc., aims at supporting the ideation process, an essential outcome of brainstorming. They enable organizations to work more efficiently and creatively in modern workspaces. Their flagship product, Quilla, is the world's largest e-writer, consisting of a portable electronic whiteboard. Being always "on" and ready to use, it displays a background document (which is imported by the user) and saves user-created annotations automatically as they are written. The purpose is to share user-created annotations for real-time, collaborative ideation. Currently, Quilla e-writers are unaware of the background document and allow only for saving and sharing the users' annotations. This proposal aims to create a new, dynamic, and flexible paradigm for enabling e-writers to process and understand user-imported documents. Computer vision methods will be explored for creating a content-aware framework that enables interactive acquisition, display, and manipulation of document, flipchart, and whiteboard images. The development of such a framework would significantly enhance the user experience and broaden Quilla's scope and functionalities. Many benefits from the proposed research will accrue to our industrial partner, and the Canadian industry, to the University of Victoria, and to the computer vision research community. While the developed algorithms will broaden the scope of business productivity tools, this project will also contribute to the strengthening of UVic's leadership role in document image analysis research, and to the training of highly qualified personnel in state of the art computer vision, machine learning, and human-computer interaction technologies.