Length Prediction of Straight Gestures Pen-based Applications


Yi-Chun Nikko Cheng

Oral Defence Date: 

Tuesday, August 15, 2006 - 14:00


TH 331


2:00 - 4:00 pm


Prof. Edward Lank, Prof. Ilmi Yoon, Prof. Dragutin Petkovic


Pen and paper had been primary recording devices in human history for more than a thousand years. However, limited researches looked into speed profile of gesture drawings until recent years. Furthermore, no study attempted to predict the length of gestures based on the characteristics of their initial acceleration. With help of advanced pen-based devices, such as PDA and Tablet PCs, this type of motion can be observed and studied via computer programs. In this thesis, we describe the design and implementation of a framework to approximate the characteristics of straight lines being dawn on pen-based devices and derive a length approximation equation based on the analysis. We present an equation that can estimate the length of a gesture by analyze its early movements with enough level of confidence.

Yi-Chun Nikko Cheng

Target prediction; stylus; tablet; length prediction; Fitts' Law; Motion; 2/3 Power Law; Minimum Jerk Law

Date Report Of Completion Filed : 09/06/2006