Applying Deep Learning in Augmented Reality Tracking
Authors:
- Omer AkgulOAOmer Akgul
- H. Ibrahim PenekliHPH. Ibrahim Penekli
- Yakup GencYGYakup Genc
- 1 January 2016
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 47-54
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.Keywords
This publication has 15 references indexed in Scilit:
- Going deeper with convolutions2015
- Caffe2014
- Learning Hierarchical Features for Scene LabelingIEEE Transactions on Pattern Analysis and Machine Intelligence, 2012
- Binary Histogrammed Intensity Patches for Efficient and Robust MatchingInternational Journal of Computer Vision, 2011
- Real-Time Detection and Tracking for Augmented Reality on Mobile PhonesIEEE Transactions on Visualization and Computer Graphics, 2009
- A performance evaluation of local descriptorsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2005
- Distinctive Image Features from Scale-Invariant KeypointsInternational Journal of Computer Vision, 2004
- Marker-less tracking for AR: a learning-based approach2002
- A Survey of Augmented RealityPRESENCE: Virtual and Augmented Reality, 1997
- Random sample consensusCommunications of the ACM, 1981