Learn More
We present a method for discovering informative patterns from data. With this method, large databases can be reduced to only a few representative data entries. Our framework encompasses also methods for cleaning databases containing corrupted data. Both on-line and off-line algorithms are proposed and experimentally checked on databases of handwritten(More)
We have designed a writer-adaptive character recognition system for on-line characters entered on a touch-terminal. It is based on a Time Delay Neural Network (TDNN) that is rst trained on examples from many writers to recognize digits and uppercase letters. The TDNN without its last layer serves as a preprocessor to an Optimal Hyperplane classi er, that(More)
This paper discusses a fairly general adaptation algorithm which augments a standard neural network to increase its recognition accuracy for a specific user . The basis for the algorithm is that the output of a neural network is characteristic of the input, even when the output is incorrect. We exploit this characteristic output by using an Output(More)
Even with highly-sensitive touchscreens and emphasis on "designing for touch", small target selection remains difficult. Good touch performance cannot solve the "fat-finger" problem, which results from occlusion and the size disparity between fingers and targets. We propose Hover Cursor, a method to improve small target selection using hover-sensing over a(More)
We describe a new circular touch gesture for scrolling called ChiralMotion<sup>TM</sup> and report on work to measure its performance. In a study using a document scrolling task, ChiralMotion outperformed linear Virtual Scrolling on a notebook computer TouchPad. Participants also indicated a preference for ChiralMotion in a follow-up questionnaire. We(More)
Force-based touch input is a novel input method for automotive interfaces. It constitutes an efficient and effective way of in-car interaction with an added degree of freedom. Research on force-based in-car interfaces is still scarce. Recommendations such as gesture sets or interaction languages that could guide future implementations have not yet been(More)
Current automotive interfaces rely heavily on touchscreen interfaces which leverage simple and intuitive direct touch interaction. Since input and output are co-located, displays have to be positioned within hand's reach. When the display is outside the reach envelope, a touchpad has been used as a control device. Current implementation of touchpads in cars(More)
With the spread of mobile device fingerprint sensors, there is an opportunity to allow interaction beyond biometric authentication on the same device. The fingerprint sensor can detect simple gestures, enabling an additional gestural input method. We conducted user research sessions with ten people to understand the usefulness of a fingerprint sensor that(More)
Steering wheel controls have become an essential interface for interacting with in-car human machine interfaces. The increase in both number and complexity of features in the car challenges the scalability of button-based mechanical steering wheel interfaces. Alternatives such as touch input are making their way into steering wheels of mass production cars.(More)
  • 1