Towards more accurate immersive 3D sketching

This doctoral work aims to reduce the gap in knowledge of how users utilize immersive 3D sketching through a better understanding of what affects them while drawing in virtual reality. My first goal is to know more about the reasons behind the reduced accuracy of 3D sketches compared to 2D ones, with an eye towards potential differences between users with different skill levels. While previous research described the various challenges of immersive 3D drawing, those descriptions have focused mostly on identifying the reasons regarding why people draw worse in 3D than 2D. In this doctoral work, my goal is to understand how the perceptual and cognitive limitations of humans affect their behaviours when working in virtual environments. The second goal of this doctoral work is to develop new user interfaces that help novice users draw better using virtual reality. I aim to allow users to express their ideas more easily, through improving stroke quality and global shape likeness, without affecting their stroke expressiveness. The user’s stroke quality measures (locally) how close a drawn stroke is to an intended one and shape likeness measures how (globally) similar a drawn object is to the intended shape. Improving both of these qualities makes sketching a useful tool to share concepts and to aid the user’s memory. My work on these two goals resulted in four different projects. Each project was previously published, and I present the full text of those four studies in this cumulative format dissertation. The four projects include 1) a study of the effect of changing the viewpoint when drawing in 3D, 2) a study of the effect of the depth perception problems of stereo displays on hand pointing in peripersonal space, 3) a system called Multiplanes, and 4) a system called Smart3DGuides. In addition to these projects, I posed a critical reflection on the user interface requirements for immersive 3D drawing systems to inform the design of future interfaces. Finally, I address this dissertation to user interface designers and HCI and design researchers who are interested in using virtual reality as a new medium to sketch