Fake Data - the Future of Advanced Computer Vision Systems

Peter Corcoran

doi:http://hdl.handle.net/10379/15508

Fake Data - the Future of Advanced Computer Vision Systems

Peter Corcoran

Electrical & Electronic Engineering

Research output: Other contribution (Published) › Other contribution

Abstract

Recent research shows that Data Augmentation techniques and Synthetic Data can improve the accuracy and reduce the susceptibility of Deep Neural Networks to Adversarial Attacks. In this presentation we consider some of the new tools that are available to build advanced virtual models that can be used to render large 2D training datasets suitable for training tomorrows advanced Computer Vision systems for deployment in consumer and smart-city use cases.

Original language	English (Ireland)
Media of output	Presentation
Publisher	bit.ly
DOIs	https://doi.org/http://hdl.handle.net/10379/15508
Publication status	Published - 1 Nov 2018

Access to Document

http://hdl.handle.net/10379/15508

Cite this

@misc{a3ff3328237644c9a1f21c3748ff1d2c,

title = "Fake Data - the Future of Advanced Computer Vision Systems",

abstract = "Recent research shows that Data Augmentation techniques and Synthetic Data can improve the accuracy and reduce the susceptibility of Deep Neural Networks to Adversarial Attacks. In this presentation we consider some of the new tools that are available to build advanced virtual models that can be used to render large 2D training datasets suitable for training tomorrows advanced Computer Vision systems for deployment in consumer and smart-city use cases. ",

author = "Peter Corcoran",

note = "The role of Augmented and Generated Data in Training Neural Networks for AI is Discussed.",

year = "2018",

month = nov,

day = "1",

doi = "http://hdl.handle.net/10379/15508",

language = "English (Ireland)",

publisher = "bit.ly",

type = "Other",

}

TY - GEN

T1 - Fake Data - the Future of Advanced Computer Vision Systems

AU - Corcoran, Peter

N1 - The role of Augmented and Generated Data in Training Neural Networks for AI is Discussed.

PY - 2018/11/1

Y1 - 2018/11/1

N2 - Recent research shows that Data Augmentation techniques and Synthetic Data can improve the accuracy and reduce the susceptibility of Deep Neural Networks to Adversarial Attacks. In this presentation we consider some of the new tools that are available to build advanced virtual models that can be used to render large 2D training datasets suitable for training tomorrows advanced Computer Vision systems for deployment in consumer and smart-city use cases.

AB - Recent research shows that Data Augmentation techniques and Synthetic Data can improve the accuracy and reduce the susceptibility of Deep Neural Networks to Adversarial Attacks. In this presentation we consider some of the new tools that are available to build advanced virtual models that can be used to render large 2D training datasets suitable for training tomorrows advanced Computer Vision systems for deployment in consumer and smart-city use cases.

U2 - http://hdl.handle.net/10379/15508

DO - http://hdl.handle.net/10379/15508

M3 - Other contribution

PB - bit.ly

ER -

Fake Data - the Future of Advanced Computer Vision Systems

Abstract

Access to Document

Fingerprint

Cite this