The advent of consumer level 3D user interface devices, e.g. Kinect or Leap Motion, have triggered a paradigm shift in the fields of user interface design and development. These devices allow the development of natural, intuitive and user-friendly interfaces that result in increased usability and enhanced user experience. A natural user interfaces (NUI) can be employed in a number of areas and applications, but with the arrival of virtual and augmented reality systems along with an ever-increasing deployment of large-scale displays, their use has become even more important. A gesture-based NUI allows the user to interact with a system in an efficient manner with little overhead in terms of technical abilities because of its intuitiveness. Therefore, it is necessary to implement and test NUIs in a number of scenarios to validate their effectiveness and robustness under different conditions. This will result in identifying the appropriate attributes of these interfaces that can lead to a better usability and user experience compared to traditional user interfaces.
1.1. Motivation and Contributions
Visualization and manipulation of 3D content, specifically 3D animated content has received a lot of attention in the past two decades. 3D animation, free-viewpoint video, or 3D video is an important area of research in the areas of Computer Vision and Computer Graphics (Carranza et al., 2003; Ye et al., 2013). The user can not only play, pause, forward, or rewind the 3D video, but can also rotate the video around any axis to watch it from any angle. In addition, the user can also scale the video to modify the appearance of the actors in the video. In principal, any 3D content viewing system that supports animation can display a 3D animation, e.g. 3dsmax, Blender, Maya etc. Traditionally, these systems employ a cursor-based user interface (GUI) that allows the user to view the 3D animation content using mouse buttons and user interface elements to manipulate the animation (DeCarlo, 2016). While these interfaces work for the small computer screens, if the 3D animation is visualized on a bigger screen where the user is visualizing from a farther distance, the control and manipulation of the 3D animation content becomes very problematic. In situations like this, a NUI for controlling and manipulating a 3D animation would prove more robust and effective.
In this paper, we present a new NUI for controlling and manipulating a 3D animation. Our system employs a gesture-based user interface using Kinect to control and manipulate a 3D animation. We design and implement a number of gestures that are detected by using the real-time skeletal data from Kinect. Using the gesture, the user can control and manipulate a 3D animation on a large screen while standing away from the display. The proposed NUI allows the user to play, pause, rewind, forward, rotate and scale the 3D animation. The system allows all forms of 3D rotations, i.e. yaw, pitch, and roll. For comparison, we also implemented a cursor-based system with traditional button-based user interface to control and manipulate the 3D animation.
We perform a very detailed user study to test the usability of NUI along with the quality of the user experience. The usability is tested in terms of its efficiency, effectiveness, simplicity, ease of learning and memorization. In addition, a number of positive and negative user experience goals are identified and measured to find the quality of user experience. Finally, we also perform a quantitative study to measure the time taken to perform a set of tasks that are used to manipulate and control a 3D animation. Our user study shows that NUI is preferred by the majority of users in terms of both the usability and user experience.
The main contributions of our work are:
- •
Gesture design for a NUI to control and manipulate a 3D animation;
- •
Implementation of a complete system that uses Kinect to detect the gestures and allows the user to interact with a 3D animation;
- •
A comprehensive user study to evaluate both the usability and user experience of the proposed NUI, both qualitatively and quantitatively.