Title: Sensing Foot Gestures from the Pocket
Author: Jeremy Scott
Publisher: UIST' 10, October 3-6, 2010 New York
Summary
Mobile interfaces on mobile phones are visually demanding. Not only that, they diminish the user experience by monopolizing the user's attention when they are focusing on another task. Because mobile phones are typically located in pockets, explicit foot movements can be used as sort of an eyes-and-hands-free input gesture for interacting with the device.
Typically interaction is done with direct touch or physical buttons. The visual demands of this experience are heavily reliant on visual feedback as output. Currently there is vibrotactile feedback that a user relies on to alert them. Additionally, sounds such as ringtones or text-tones alert the user of messages without needing visual feedback. The authors want to rely on the 3-axis accelerometer found in iPhones to determine what sort of foot movements were made.
First they studied foot-based interaction space. They looked at four distinct actions: dorsiflexion, plantar flexion, heel rotation, and toe rotation. They found that users were more capable of consistent plantar flexion than dorsiflexion. Additionally, heel rotations were more comfortable than toe rotations.
Next they placed an iPhone in 3 different locations: front pocket, attached to the hip, and side pocket. They asked the users to perform the foot gestures. The side pocket had the best accuracy (~86%), with the front pocket showing the next best action accuracy (~60%).
Discussion
I thought that this study was absolutely ridiculous. I don't see foot or leg actions to be practical whatsoever when it comes to interacting with mobile phones. When would someone ever need to perform some action on their phone when it's in their pocket without a) looking at it and b) with their foot? The only thing that I can think of is if it rings and you want to cancel the ring. But even then, usually if you want to send the call to voicemail you're probably not in a position to answer it and thus, not in a position that making some kind of leg twitch be a completely normal thing to do.
Maybe this might be applicable to some workout apps or something along those lines. I feel like the authors of this article could have laid out a few more times when their research might have been practical besides someone needing to switch songs while standing at a bus stop. All iPhones come with a set of headphones with a button that you can click to pause or change songs already.
Image: from the article.
Typically interaction is done with direct touch or physical buttons. The visual demands of this experience are heavily reliant on visual feedback as output. Currently there is vibrotactile feedback that a user relies on to alert them. Additionally, sounds such as ringtones or text-tones alert the user of messages without needing visual feedback. The authors want to rely on the 3-axis accelerometer found in iPhones to determine what sort of foot movements were made.
First they studied foot-based interaction space. They looked at four distinct actions: dorsiflexion, plantar flexion, heel rotation, and toe rotation. They found that users were more capable of consistent plantar flexion than dorsiflexion. Additionally, heel rotations were more comfortable than toe rotations.
Next they placed an iPhone in 3 different locations: front pocket, attached to the hip, and side pocket. They asked the users to perform the foot gestures. The side pocket had the best accuracy (~86%), with the front pocket showing the next best action accuracy (~60%).
Discussion
I thought that this study was absolutely ridiculous. I don't see foot or leg actions to be practical whatsoever when it comes to interacting with mobile phones. When would someone ever need to perform some action on their phone when it's in their pocket without a) looking at it and b) with their foot? The only thing that I can think of is if it rings and you want to cancel the ring. But even then, usually if you want to send the call to voicemail you're probably not in a position to answer it and thus, not in a position that making some kind of leg twitch be a completely normal thing to do.
Maybe this might be applicable to some workout apps or something along those lines. I feel like the authors of this article could have laid out a few more times when their research might have been practical besides someone needing to switch songs while standing at a bus stop. All iPhones come with a set of headphones with a button that you can click to pause or change songs already.
Image: from the article.
I agree, this was a bizarre paper. It would have been much better for them to research something along the lines of a workout pedometer.
ReplyDelete