top of page

explore uncc

Gesture-based Interactive System to Explore University Information 

System Description

We designed a set of gesture commands as a model for a gesture-based interactive information system with large screen displays in public locations that provide information about an institution. This walk-up-and-use information display was developed in the context of a university department and serves as a framework for an interactive display that engages passersby to easily explore various entities and relations about courses and faculty on a public display. The walk-up-and-use information display utilizes gesture recognition via Microsoft Kinect and is designed so that multiple users may interact with the information system in a public location. The design goals are to design a discoverable interaction system for mid-air gestures, to design an engaging and enjoyable experience on a public display, and to enhance gesture recognition in a university setting.

 

My Role: In order to determine which gestures are the most easily discovered for interaction design, a think-aloud gesture-elicitation study was conducted to come up with common intuitive gestures defined by the users. I designed user interface and conducted user study and analysis. 

Design Goal

This system was successful in interacting with information systems with gestures and is an engaging visual interface. It has proven to be an attractive system that either draws users on its own or users bring other users to the system through word of mouth based on their own experience of interacting with the system. The design goal is described as follows.

  • Recognizing Intention: Our application must improve upon the previous iteration’s ability to recognize the intent of the user.

  • Accessible Navigation: We should strive to discover innovative new ways of navigating through the information displayed, making our application more accessible and easy to use.

  • Engaging Interaction: In order for our application to be truly effective, students and other potential users must be enticed to interact with the program as they pass by.

System Design

The interface we have developed employs flock-based clustering to both organize information with respect to content and to provide visual representations. Information visualization provides users facilities to gain an overview of the entire information, and filter information selectively for more detail.

Initial Design

Picture1.png
Picture2.jpg

Final Design

Picture3.png

To provide clarity and visual hierarchy, previous system used color to differentiate between various information types. This information can be arranged by type or discipline in order to provide users with multiple ways to explore the relationships between various elements of the system. We designed the systems prototype, focusing on the Software and Information Systems (SIS) department in the College of Computing and Informatics (CCI) at University of North Carolina Charlotte (UNCC), to display information about five different disciplines within the SIS department. This system served as the framework to provide an interactive display for students, faculty, and visitors to easily explore these types of relationships. These relationships include, but are not limited to, classes, professors, research topics, organizations, and events both within and across departments at the university. Information was presented to the user in the form of 15-second video clips, images, and text descriptions.

User Study

The first study involves a heuristic evaluation study. The second study involves the usability evaluation of the system. The usability will be measured based on the participants’ experience after they perform a set of tasks to explore the system.

Heuristic Evaluation

We identified the following issues: 

  • Visibility of system status

  • User control and freedom

  • Recognition rather than recall

  • Flexibility and efficiency of use

  • Aesthetic and minimalist design

Usability and Attractiveness Evaluation Study

 

We found that while our initial design was attractive, there were several ergonomic issues that he felt users would immediately encounter. There was difficulty in choosing from a linear horizontal row of buttons across the bottom or top of the screen, as the hand moved out of the Kinect’s established tracking threshold when reaching upwards or downwards to the corners of the screen. This observation led to the arrangement of category bubbles in an arc around the center of the screen, which made them much easier to reach without error.

Some users even assumed that the circular photos of faculty would be buttons as well, and were disappointed to not be rewarded with more information after hovering over these images.

While the second option seemed less cluttered and more immediately advantageous, it seems that this would confuse the user as there would be less of a connection between the selected course bubble and its related information.

bottom of page