Affiliation:
1. University of British Columbia, Vancouver, Canada
Abstract
Touch sensing on ad-hoc surfaces has the potential to transform everyday surfaces in the environment - desks, tables and walls - into tactile, touch-interactive surfaces, creating large, comfortable interactive spaces without the cost of large touch sensors. Depth sensors are a promising way to provide touch sensing on arbitrary surfaces, but past systems have suffered from high latency and poor touch detection accuracy. We apply a novel state machine-based approach to analyzing touch events, combined with a machine-learning approach to predictively classify touch events from depth data with lower latency and higher touch accuracy than previous approaches. Our system can reduce end-to-end touch latency to under 70ms, comparable to conventional capacitive touchscreens. Additionally, we open-source our dataset of over 30,000 touch events recorded in depth, infrared and RGB for the benefit of future researchers.
Funder
Natural Sciences and Engineering Research Council of Canada
Publisher
Association for Computing Machinery (ACM)
Subject
Computer Networks and Communications,Human-Computer Interaction,Social Sciences (miscellaneous)
Reference34 articles.
1. Fingertip detection approach on depth image sequences for interactive projection system
2. A Predictive Approach for an End-to-End Touch-Latency Measurement
3. Reducing Latency with a Continuous Prediction
4. Zhi Chai and Roy Shilkrot . 2018. Enhanced Touchable Projector-depth System with Deep Hand Pose Estimation. arXiv:1812.11090 [cs] , Dec., arxiv:1812.11090 arXiv : 1812 .11090 Zhi Chai and Roy Shilkrot. 2018. Enhanced Touchable Projector-depth System with Deep Hand Pose Estimation. arXiv:1812.11090 [cs], Dec., arxiv:1812.11090 arXiv: 1812.11090
5. Hammer Time!
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献