Abstract

Humans use gestures to emphasize ideas and disseminate information. Their importance is apparent in how we continuously augment social interactions with motion—gesticulating in harmony with nearly every utterance to ensure observers understand that which we wish to communicate, and their relevance has not escaped the HCI community's attention. For almost as long as computers have been able to sample human motion at the user interface boundary, software systems have been made to understand gestures as command metaphors. Customization, in particular, has great potential to improve user experience, whereby users map specific gestures to specific software functions. However, custom gesture recognition remains a challenging problem, especially when training data is limited, input is continuous, and designers who wish to use customization in their software are limited by mathematical attainment, machine learning experience, domain knowledge, or a combination thereof. Data collection, filtering, segmentation, pattern matching, synthesis, and rejection analysis are all non-trivial problems a gesture recognition system must solve. To address these issues, we introduce The Dollar General (TDG), a complete pipeline composed of several novel continuous custom gesture recognition techniques. Specifically, TDG comprises an automatic low-pass filter tuner that we use to improve signal quality, a segmenter for identifying gesture candidates in a continuous input stream, a classifier for discriminating gesture candidates from non-gesture motions, and a synthetic data generation module we use to train the classifier. Our system achieves high recognition accuracy with as little as one or two training samples per gesture class, is largely input device agnostic, and does not require advanced mathematical knowledge to understand and implement. In this dissertation, we motivate the importance of gestures and customization, describe each pipeline component in detail, and introduce strategies for data collection and prototype selection.

Notes

If this is your thesis or dissertation, and want to learn how to access it or for more information about readership statistics, contact us at STARS@ucf.edu

Graduation Date

2020

Semester

Fall

Advisor

Laviola II, Joseph

Degree

Doctor of Philosophy (Ph.D.)

College

College of Engineering and Computer Science

Department

Computer Science

Degree Program

Computer Science

Format

application/pdf

Identifier

CFE0008391; DP0023828

URL

https://purls.library.ucf.edu/go/DP0023828

Language

English

Release Date

December 2020

Length of Campus-only Access

None

Access Status

Doctoral Dissertation (Open Access)

Share

COinS