All geek questions in one place

Manual use detection from device

Is there any evidence that we could reference to determine if a person is using a device (smartphone / tablet) with his left or right hand?

My guess is that you can use the accelerometer data to detect a slight tilt, perhaps only when the user is manipulating somehow on the screen.

The answer I'm looking for will say something like: "Studies show that 90% of right-handed users who use the input mechanism will tilt their phone an average of 5 ° when entering data, and 90% of the users on the left use the mechanism input has its own phone, tilted by an average of -5 °. "

Having this data, one could read the accelerometer data and be able to make informed decisions regarding the placement of elements on the screen that otherwise might be in the way for left-handed users or right-handed people.

+8
source share
2 answers

You can definitely do this, but if it were me, I would try a less complicated approach. First you need to admit that no specific approach will give 100% accurate results - they will be speculations, but hopefully very likely. With that said, I would explore simple capture points for basic touch events. You can use these data points and pull the x / y axis when touching the start / end:

touchStart: , , .

touchEnd: , .

- , , / . , , , , , . , touchStart X , touchEnd X, , . - , touchStart X , touchEnd X, , . . :

enter image description here

enter image description here

. !

http://www.javascriptkit.com/javatutors/touchevents.shtml

+8

, . , 2012-2016. , . , , , , .

: http://huyle.de/wp-content/papercite-data/pdf/le2019investigating.pdf

, Python 3.6 , , , MIT. Android.

Git : https://github.com/interactionlab/CapFingerId

+1

Source: https://habr.com/ru/post/1569532/

More articles:


All Articles