Affiliation:
1. University of Central Florida, Orlando, FL
Abstract
Sketch recognition has the potential to be an important input method for computers in the coming years, particularly for STEM (science, technology, engineering, and math) education. However, designing and building an accurate and sophisticated sketch recognition system is a time-consuming and daunting task. Since sketch recognition mistakes are still common, it is important to understand how users perceive and tolerate recognition errors and other user interface elements with these imperfect systems. In order to solve this problem, we developed a Wizard of Oz sketch recognition tool, the WOZ Recognizer, that supports controlled recognition accuracy, multiple recognition modes, and multiple sketching domains for performing controlled experiments. We present the design of the WOZ Recognizer and our process for representing recognition domains using graphs and symbol alphabets. In addition, we discuss how sketches are altered, how to control the WOZ Recognizer, and how users interact with it. Finally, we present an expert user case study that examines the WOZ Recognizer’s usability.
Publisher
Association for Computing Machinery (ACM)
Subject
Artificial Intelligence,Human-Computer Interaction
Cited by
3 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献