Comment 1
Comment 2
References:
Title: Sensing Foot Gestures from the Pocket
Authors: Jeremy Scott, David Dearman, Koji Yatani, and Khai N Truong
Venue: UIST 2010, Oct 3-6, 2010
Summary:
In this paper, the authors describe a method of input for mobile phones using the foot of the user. In order to create a concept, they performed a study to find out what gestures have the highest accuracy and are the most comfortable to the user using the apparatus to the right. They discovered that heel rotation, flexing at the toe, and double-tapping the foot were the most accurate and comfortable.Then, they created an application for the iPhone to recognize these gestures, and then tested their accuracy from different locations on the body. They found that on a holster on the side and in a side pocket are the most accurate, after a small machine learning period. They then describe the limitations of the current system, including differentiating running from the double-tap and keeping the gestures accurate if the phone moves around in the user's pocket.
Discussion:
This paper, while scientifically strong and with an original concept, didn't interest me all that much. The uses for foot gestures don't seem readily apparent to me; additionally, I don't think it will be easy to determine inputs that the user means to perform versus innocuous activities like walking. I feel that other alternative input methods that they listed early in the paper such as speech input or rear buttons would probably work better than this one.
While I agree with you that other approaches like speech input seem better options for what they're trying to do, I do think it's cool they did a user study before they developed their application. In trying to thinking of uses for this particular system, perhaps it could be used as a pedometer.
ReplyDeleteI don't see people ever trying to use their feet to interact with things, other than something to step on or pressure sensitive. It's just too awkward. I didn't enjoy this article either.
ReplyDelete