ABSTRACT
The latency of current mobile devices' touchscreens is around 100ms and has widely been explored. Latency down to 2ms is noticeable, and latency as low as 25ms reduces users' performance. Previous work reduced touch latency by extrapolating a finger's movement using an ensemble of shallow neural networks and showed that predicting 33ms into the future increases users' performance. Unfortunately, this prediction has a high error. Predicting beyond 33ms did not increase participants' performance, and the error affected the subjective assessment. We use more recent machine learning techniques to reduce the prediction error. We train LSTM networks and multilayer perceptrons using a large data set and regularization. We show that linear extrapolation causes an 116.7% higher error and the previously proposed ensembles of shallow networks cause a 26.7% higher error compared to the LSTM networks. The trained models, the data used for testing, and the source code is available on GitHub.
- Glen Anderson, Rina Doherty, and Subhashini Ganapathy. 2011. User Perception of Touch Screen Latency. In Design, User Experience, and Usability. Theory, Methods, Tools and Practice. Springer, 195--202.Google Scholar
- François Bérard and Renaud Blanch. 2013. Two Touch System Latency Estimators: High Accuracy and Low Overhead. In Proceedings of the 2013 ACM International Conference on Interactive Tabletops and Surfaces (ITS '13). ACM, New York, NY, USA, 241--250. Google ScholarDigital Library
- Elie Cattan, Amélie Rochet-Capellan, Pascal Perrier, and François Bérard. 2015. Reducing Latency with a Continuous Prediction: Effects on Users' Performance in Direct-Touch Target Acquisitions. In Proceedings of the 2015 International Conference on Interactive Tabletops & Surfaces (ITS '15). ACM, New York, NY, USA, 205--214. Google ScholarDigital Library
- Elie Cattan, Amélie Rochet-Capellan, Pascal Perrier, and François Bérard. 2017. Does Practice Make Perfect?. In Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems (CHI '17). ACM, New York, NY, USA, 5619--5629. Google ScholarDigital Library
- Jonathan Deber, Bruno Araujo, Ricardo Jota, Clifton Forlines, Darren Leigh, Steven Sanders, and Daniel Wigdor. 2016. Hammer Time!: A Low-Cost, High Precision, High Accuracy Tool to Measure the Latency of Touchscreen Devices. In Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems (CHI '16). ACM, New York, NY, USA, 2857--2868. Google ScholarDigital Library
- Xavier Glorot and Yoshua Bengio. 2010. Understanding the difficulty of training deep feedforward neural networks. In Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics (Proceedings of Machine Learning Research), Yee Whye Teh and Mike Titterington (Eds.), Vol. 9. PMLR, Chia Laguna Resort, Sardinia, Italy, 249--256. http://proceedings.mir.press/v9/glorot10a.htmlGoogle Scholar
- Niels Henze, Markus Funk, and Alireza Sahami Shirazi. 2016. Software-reduced Touchscreen Latency. In Proceedings of the 18th International Conference on Human-Computer Interaction with Mobile Devices and Services (MobileHCI '16). ACM, New York, NY, USA, 434--441. Google ScholarDigital Library
- Niels Henze and Martin Pielot. 2013. App Stores: External Validity for Mobile HCI. interactions 20, 2 (March 2013), 33--38. Google ScholarDigital Library
- Niels Henze, Alireza Sahami Shrazi, Albrecht Schmidt, Martin Pielot, and Florian Michahelles. 2013. Empirical Research through Ubiquitous Data Collection. Computer 46, 6 (June 2013), 74--76. Google ScholarDigital Library
- Sepp Hochreiter and Jürgen Schmidhuber. 1997. Long short-term memory. Neural computation 9, 8 (1997), 1735--1780. Google ScholarDigital Library
- Ricardo Jota, Albert Ng, Paul Dietz, and Daniel Wigdor. 2013. How fast is fast enough?: a study of the effects of latency in direct-touch pointing tasks. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. ACM, 2291--2300. Google ScholarDigital Library
- Topi Kaaresoja and Stephen Brewster. 2010. Feedback is... late: measuring multimodal delays in mobile device touchscreen interaction. In International Conference on Multimodal Interfaces and the Workshop on Machine Learning for Multimodal Interaction. ACM, 2. Google ScholarDigital Library
- Diederik P. Kingma and Jimmy Ba. 2014. Adam: A Method for Stochastic Optimization. CoRR abs/1412.6980 (2014). http://arxiv.org/abs/1412.6980Google Scholar
- Yann LeCun, Yoshua Bengio, and Geoffrey Hinton. 2015. Deep learning. Nature 521, 7553 (2015), 436--444.Google Scholar
- Albert Ng, Michelle Annett, Paul Dietz, Anoop Gupta, and Walter F. Bischof. 2014. In the Blink of an Eye: Investigating Latency Perception During Stylus Interaction. In Proceedings of the 32Nd Annual ACM Conference on Human Factors in Computing Systems (CHI '14). ACM, New York, NY, USA, 1103--1112. Google ScholarDigital Library
- Albert Ng, Julian Lepinski, Daniel Wigdor, Steven Sanders, and Paul Dietz. 2012. Designing for low-latency direct-touch input. In Proceedings of the 25th annual ACM symposium on User interface software and technology. ACM, 453--464. Google ScholarDigital Library
- Sebastian Ruder. 2016. An overview of gradient descent optimization algorithms. CoRR abs/1609.04747 (2016). http://arxiv.org/abs/1609.04747Google Scholar
Index Terms
- Improving software-reduced touchscreen latency
Recommendations
Software-reduced touchscreen latency
MobileHCI '16: Proceedings of the 18th International Conference on Human-Computer Interaction with Mobile Devices and ServicesDevices with touchscreens have an inherent latency. When a user's finger drags an object across the screen the object follows with a latency of around 100ms for current devices. Previous work showed that latencies down to 25ms reduce users' performance ...
PredicTouch: A System to Reduce Touchscreen Latency using Neural Networks and Inertial Measurement Units
ISS '17: Proceedings of the 2017 ACM International Conference on Interactive Surfaces and SpacesTouchscreens 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 ...
Next-Point Prediction Metrics for Perceived Spatial Errors
UIST '16: Proceedings of the 29th Annual Symposium on User Interface Software and TechnologyTouch screens have a delay between user input and corresponding visual interface feedback, called input 'latency' (or 'lag'). Visual latency is more noticeable during continuous input actions like dragging, so methods to display feedback based on the ...
Comments