by: Jeremy Scott, David Dearman, Koji Yatani, and Khai N. Truong
Authors
Scott did research for this project as an undergrad at the University of Toronto, and now attends MIT as a graduate student. Dearman seems nonexistent but clearly has associations with the U of T like everyone else, possibly as a PhD candidate like Yatani, or another professor like Truong.
Presentation Venue
"UIST '10 Proceedings of the 23nd annual ACM symposium on User interface software and technology" as per the document specs from the ACM digital library, dl.acm.org. The presentation took place in New York City, NY, 2010. (They think they can, they think they can...)
Summary
The title of this paper mostly sums it up. It is about using foot gestures to perform actions on a mobile device in a pocket or holster at the hip.
Hypothesis
Scott hypothesizes that gestures performed by a foot are viable modes of input for mobile devices, and that they save precious screen space in addition to not consuming the user's entire attention for performing those actions.
Methods
The study was performed using mobile devices with standard accelerometers, motion capture cameras, and a normalized foot model. Users wore the normalized foot model to account for people with larger/smaller feet than other participants. Motion capture helped the researchers to monitor and measure their data, while the accelerometer in the mobile device did the actual work of sensing gestures performed to decide what action or selection is supposed to be performed. Participants were instructed to make various types of selections with these gestures.
Results
Gesture search found that most gestures were accurate to within about 10 degrees, but of the four gestures, participants preferred only two, heel rotation, and lifting the heel. The device seemed to be more accurate when placed in the side holster than when placed in the front or back pocket of the participant, and overall it seemed able to correctly determine the action about 80% of the time.
Discussion
First off, does this make anyone else think of the April Fools joke played by Google this year? In theory, the idea has potential. We save screen space, as they desired to accomplish. The gestures do not require looking at the screen at all, and can be performed without even taking the device out of the pocket. But, say you're walking down the street and you see someone pause and just slightly lift his leg... All that needs to happen now is for said person to wave his hand behind him, and you know he just let loose a present for other passerby's noses. But wait, that's just a gesture! Or, that guy over there paused mid disco dance... wait! that's a gesture, too! I think these gestures are unnatural because they do not take into account whether or not people would actually feel comfortable making such body language in public settings. It also does not account for the tendency for people to move along not wanting to be noticed, rather than acting rather animated wherever they happen to be standing. Hence, my comment on Gmail actions. As a slightly more serious question, do we all have to wear foot normalizers for this to work?
No comments:
Post a Comment