Touchscreens are the dominant input mechanism for a variety of devices. One of the main limitations of touchscreens is the latency to receive input, refresh, and respond. This latency is easily perceivable and reduces users' performance. Previous work proposed to reduce latency by extrapolating finger movements to identify future movements - albeit with limited success. In this paper, we propose PredicTouch, a system that improves this extrapolation using inertial measurement units (IMUs). We combine IMU data with users' touch trajectories to train a multi-layer feedforward neural network that predicts future trajectories. We found that this hybrid approach (software: prediction, and hardware: IMU) can significantly reduce the prediction error, reducing latency effects. We show that using a wrist-worn IMU increases the throughput by 15% for finger input and 17% for a stylus.
%0 Conference Paper
%1 conf/tabletop/LeSGH17
%A Le, Huy Viet
%A Schwind, Valentin
%A Göttlich, Philipp
%A Henze, Niels
%B Proceedings of the ACM International Conference on Interactive Surfaces and Spaces (ISS)
%D 2017
%E Subramanian, Sriram
%E Steimle, Jürgen
%E Dachselt, Raimund
%E Plasencia, Diego Martínez
%E Grossman, Tovi
%I ACM
%K from:leonkokkoliadis sfbtrr161 2017 C04
%P 230-239
%R 10.1145/3132272.3134138
%T PredicTouch: A System to Reduce Touchscreen Latency using Neural Networks and Inertial Measurement Units
%U https://doi.org/10.1145/3132272.3134138
%X Touchscreens are the dominant input mechanism for a variety of devices. One of the main limitations of touchscreens is the latency to receive input, refresh, and respond. This latency is easily perceivable and reduces users' performance. Previous work proposed to reduce latency by extrapolating finger movements to identify future movements - albeit with limited success. In this paper, we propose PredicTouch, a system that improves this extrapolation using inertial measurement units (IMUs). We combine IMU data with users' touch trajectories to train a multi-layer feedforward neural network that predicts future trajectories. We found that this hybrid approach (software: prediction, and hardware: IMU) can significantly reduce the prediction error, reducing latency effects. We show that using a wrist-worn IMU increases the throughput by 15% for finger input and 17% for a stylus.
%@ 978-1-4503-4691-7
@inproceedings{conf/tabletop/LeSGH17,
abstract = {Touchscreens are the dominant input mechanism for a variety of devices. One of the main limitations of touchscreens is the latency to receive input, refresh, and respond. This latency is easily perceivable and reduces users' performance. Previous work proposed to reduce latency by extrapolating finger movements to identify future movements - albeit with limited success. In this paper, we propose PredicTouch, a system that improves this extrapolation using inertial measurement units (IMUs). We combine IMU data with users' touch trajectories to train a multi-layer feedforward neural network that predicts future trajectories. We found that this hybrid approach (software: prediction, and hardware: IMU) can significantly reduce the prediction error, reducing latency effects. We show that using a wrist-worn IMU increases the throughput by 15% for finger input and 17% for a stylus.},
added-at = {2020-03-11T16:07:24.000+0100},
author = {Le, Huy Viet and Schwind, Valentin and Göttlich, Philipp and Henze, Niels},
biburl = {https://puma.ub.uni-stuttgart.de/bibtex/2a6e10b2f1214c9a943b6560ddaafce1e/sfbtrr161},
booktitle = {Proceedings of the ACM International Conference on Interactive Surfaces and Spaces (ISS)},
doi = {10.1145/3132272.3134138},
editor = {Subramanian, Sriram and Steimle, Jürgen and Dachselt, Raimund and Plasencia, Diego Martínez and Grossman, Tovi},
ee = {https://doi.org/10.1145/3132272.3134138},
interhash = {ae3ebae01b2113c09d016da486d26f10},
intrahash = {a6e10b2f1214c9a943b6560ddaafce1e},
isbn = {978-1-4503-4691-7},
keywords = {from:leonkokkoliadis sfbtrr161 2017 C04},
pages = {230-239},
publisher = {ACM},
timestamp = {2020-03-11T15:07:24.000+0100},
title = {PredicTouch: A System to Reduce Touchscreen Latency using Neural Networks and Inertial Measurement Units},
url = {https://doi.org/10.1145/3132272.3134138},
year = 2017
}