Probabilistic Palm Rejection Using Spatiotemporal Touch Features and Iterative Classification (CHI ’14)
Tablet computers are often called upon to emulate classical pen-and-paper input. However, most touch devices today lack palm rejection features – most notably the highly popular Apple iPad tablets. Failure to reject palms effectively in a pen or touch input system results in ergonomic issues, accidental activation and unwanted inputs, precluding fluid and efficient use of these input systems. This issue has been well explored in the academic literature (see paper for a review)
We present a probabilistic touch filtering approach that uses the temporal evolution of touch contacts to reject palms. Our system improves upon previous approaches, reducing accidental palm inputs to 0.016 per pen stroke, while correctly passing 98% of stylus inputs. Further, our system requires no initial configuration and is independent of screen orientation and user handedness. We review contemporary palm rejection implementations and compare our approach against two applications in a user study, offering the first publicly available comparison of such systems.
Schwarz, J., Xiao, R., Mankoff, J., Hudson, S., and Harrison, C. Probabilistic Palm Rejection Using Spatiotemporal Touch Features and Iterative Classification. In Proceedings of the 32nd Annual SIGCHI Conference on Human Factors in Computing Systems (Toronto, Canada, April 26 – May 1, 2014). CHI ’14. ACM, New York, NY. 2009-2012.