Presentation 2009-10-08
Camera Tracking Using LLAH for Document Based Augmented Reality Applications
Hideaki UCHIYAMA, Hideo SAITO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we propose a method for text document tracking based on on-line learning of the local arrangement of keypoints using LLAH. Locally Likely Arrangement Hashing (LLAH) is a keypoint retrieval method by indexing local arrangement of neighbor keypoints for document image retrieval. Since LLAH is available in case of nearly top view, we extend LLAH to free viewpoint tracking by learning the behavior of keypoint arrangement. Our method starts document tracking from a top view. If the new patterns of keypoints appear while moving the camera, the database of local arrangement is incrementally updated to reflect these new patterns. Thanks to the update, our system can detect the document over a large range of viewpoints.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Tracking / Registration / LLAH / Augmented Reality
Paper # MVE2009-39
Date of Issue

Conference Information
Committee MVE
Conference Date 2009/10/1(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Media Experience and Virtual Environment (MVE)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Camera Tracking Using LLAH for Document Based Augmented Reality Applications
Sub Title (in English)
Keyword(1) Tracking
Keyword(2) Registration
Keyword(3) LLAH
Keyword(4) Augmented Reality
1st Author's Name Hideaki UCHIYAMA
1st Author's Affiliation Keio University()
2nd Author's Name Hideo SAITO
2nd Author's Affiliation Keio University
Date 2009-10-08
Paper # MVE2009-39
Volume (vol) vol.109
Number (no) 215
Page pp.pp.-
#Pages 6
Date of Issue