Applying deep learning in augmented reality tracking

Files

Applying Deep Learning in Augmented Reality Tracking.pdf (848.13 KB)

Date

2016-11-12

Authors

Editor(s)

Advisor

Supervisor

Co-Advisor

Co-Supervisor

Instructor

Source Title

Proceedings - 12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016

Print ISSN

Electronic ISSN

Publisher

IEEE

Volume

Issue

Pages

47 - 54

Language

English

Type

Conference Paper

Journal Title

Journal ISSN

Volume Title

Usage Stats
3
views
86
downloads

Attention Stats

Series

Abstract

An existing deep learning architecture has been adapted to solve the detection problem in camera-based tracking for augmented reality (AR). A known target, in this case a planar object, is rendered under various viewing conditions including varying orientation, scale, illumination and sensor noise. The resulting corpus is used to train a convolutional neural network to match given patches in an incoming image. The results show comparable or better performance compared to state of art methods. Timing performance of the detector needs improvement but when considered in conjunction with the robust pose estimation process promising results are shown. © 2016 IEEE.

Course

Other identifiers

Book Title

Keywords

Augmented reality, Neural networks, Convolutional neural network, Detection problems, Learning architectures, Marker detections, Marker tracker, State-of-art methods, Timing performance, Viewing conditions, Deep learning

Degree Discipline

Degree Level

Degree Name

Citation

Permalink

http://hdl.handle.net/11693/37607

Published Version (Please cite this version)

https://doi.org/10.1109/SITIS.2016.17

Collections

Scholarly Publications - Computer Engineering