Effective 2D Stroke-based Gesture Augmentation for RNNs
Open Access
- 19 April 2023
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1-13
Abstract
Recurrent neural networks (RNN) require large training datasets from which they learn new class models. This limitation prohibits their use in custom gesture applications where only one or two end user samples are given per gesture class. One common way to enhance sparse datasets is to use data augmentation to synthesize new samples. Although there are numerous known techniques, they are often treated as standalone approaches when in reality they are often complementary. We show that by intelligently chaining augmentation techniques together that simulate different gesture production variability types, such as those affecting the temporal and spatial qualities of a gesture, we can significantly increase RNN accuracy without sacrificing training time. Through experimentation on four public stroke-based 2D gesture datasets, we show that RNNs trained with our data augmentation chaining technique achieves state-of-the-art recognition accuracy in both writer-dependent and writer-independent test scenarios.Keywords
This publication has 64 references indexed in Scilit:
- Handwriting synthesis: classifications and techniquesInternational Journal on Document Analysis and Recognition (IJDAR), 2014
- Relative accuracy measures for stroke gestures2013
- Memorability of pre-designed and user-defined gesture sets2013
- Gestures as point clouds2012
- Rejection of non-meaningful activities for HMM-based activity recognition systemImage and Vision Computing, 2010
- Modeling Shapes for Pattern Recognition: A Simple Low-Cost Spline-based ApproachAdvances in Electrical and Computer Engineering, 2008
- Investigating user tolerance for errors in vision-enabled gesture-based interactions2006
- Off-line, handwritten numeral recognition by perturbation methodIEEE Transactions on Pattern Analysis and Machine Intelligence, 1997
- A kinematic theory of rapid human movementsBiological Cybernetics, 1995
- Specifying gestures by exampleACM SIGGRAPH Computer Graphics, 1991