Abstract
An unintentional hand drift adversely affects the typing performance of conventional virtual keyboards. To overcome this, we proposed to utilize the typing patterns of skilled typists. First, as most typists enter the keys in the same column with a predetermined finger only, we restricted these keys to be typed by their corresponding fingers. Second, our investigation of skilled typists demonstrated that hand poses vary when the typists touch different keys. Thus, rather than locating the touch point as in the case of existing virtual keyboards, we attempted to use unique hand poses to infer the target key. Based on these two techniques, we implemented a novel hand poses aware virtual keyboard that is tolerant of hand drift. Our experimental studies yielded the following results: 1) most of the QWERTY-familiar typists who have varying typing habits were easily adaptable to the proposed keyboard design and 2) the proposed keyboard outperformed existing virtual keyboards in terms of typing speed and several error rates, and eventually achieved a typing speed of approximately 56 WPM.
Highlights
Despite the prevalence of touchscreens, ten-finger typing on virtual keyboards is slower and less accurate than physical keyboards [8]
We decided that the principal component analysis (PCA) outputted 25 components and the Multi-layer perceptron (MLP) was trained with the sigmoid function as its activation function and 60 neurons in its hidden layer
For the two index fingers, whose target keys are the most difficult to infer compared to those of the other fingers, Figure 7 shows the learning curves of the MLP models whenever the original hand joint vector was transformed by each step, including the Normalizer, the StandardScaler, and the PCA
Summary
Despite the prevalence of touchscreens, ten-finger typing on virtual keyboards is slower and less accurate than physical keyboards [8]. We referred to this technique as Key Pre-allocation. To interpret typists’ intentions more precisely, we propose to infer a target key based on hand poses. We referred to this technique as Key Inference based on Hand Poses. As hand pose based inference is independent of where the hand is located, this technique further allows the key inference to be tolerant towards hand drift These two techniques allow us to implement a novel virtual keyboard that reduces typing errors and tolerates hand drift as well. We will discuss how to apply our techniques to other skilled typists who experience difficulty in adapting to Key Pre-allocation
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
