Leveraging Gans to Improve Continuous Path Keyboard Input Models

Mehra, Akash; Bellegarda, Jerome R.; Bapat, Ojas; Lal, Partha; Wang, Xin

doi:10.1109/icassp40776.2020.9052978

Cited by 10 publications

(1 citation statement)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In speech processing, GANs have been applied to speech synthesis [17], acoustic scene classification [26], and speech recognition [16]. GANs have also been used to generate synthetic path data to improve the performance of a continuous path keyboard [25]. To our knowledge, there is no prior work investigating using GANs to synthesize skeleton data for data augmentation.…”

Section: Related Workmentioning

confidence: 99%

The Imaginative Generative Adversarial Network: Automatic Data Augmentation for Dynamic Skeleton-Based Hand Gesture and Human Action Recognition

Shen¹,

Dudley²,

Kristensson³

2021

Preprint

View full text Add to dashboard Cite

Deep learning approaches deliver state-of-the-art performance in recognition of spatiotemporal human motion data. However, one of the main challenges in these recognition tasks is limited available training data. Insufficient training data results in over-fitting and data augmentation is one approach to address this challenge. Existing data augmentation strategies, such as transformations including scaling, shifting and interpolating, require hyperparameter optimization that can easily cost hundreds of GPU hours. In this paper, we present a novel automatic data augmentation model, the Imaginative Generative Adversarial Network (GAN) that approximates the distribution of the input data and samples new data from this distribution. It is automatic in that it requires no data inspection and little hyperparameter tuning and therefore it is a lowcost and low-effort approach to generate synthetic data. The proposed data augmentation strategy is fast to train and the synthetic data leads to higher recognition accuracy than using data augmented with a classical approach.

show abstract

Section: Related Workmentioning

confidence: 99%