On-line document registering and retrieving system for AR annotation overlay

Hideaki Uchiyama, Julien Pilet, Hideo Saito

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

We propose a system that registers and retrieves text documents to annotate them on-line. The user registers a text document captured from a nearly top view and adds virtual annotations. When the user thereafter captures the document again, the system retrieves and displays the appropriate annotations, in real-time and at the correct location. Registering and deleting documents is done by user interaction. Our approach relies on LLAH, a hashing based method for document image retrieval. At the on-line registering stage, our system extracts keypoints from the input image and stores their descriptors computed from their neighbors. After registration, our system can quickly find the stored document corresponding to an input view by matching keypoints. From the matches, our system estimates the geometrical relationship between the camera and the document for accurately overlaying the annotations. In the experimental results, we show that our system can achieve on-line and real-time performances.

Original languageEnglish
Title of host publicationACM International Conference Proceeding Series
DOIs
Publication statusPublished - 2010
Event1st Augmented Human International Conference, AH'10 - Megeve, France
Duration: 2010 Apr 22010 Apr 3

Other

Other1st Augmented Human International Conference, AH'10
CountryFrance
CityMegeve
Period10/4/210/4/3

Fingerprint

Image retrieval
Cameras

Keywords

  • augmented reality
  • document retrieval
  • feature matching
  • LLAH
  • Poes estimation

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Software

Cite this

Uchiyama, H., Pilet, J., & Saito, H. (2010). On-line document registering and retrieving system for AR annotation overlay. In ACM International Conference Proceeding Series [1785478] https://doi.org/10.1145/1785455.1785478

On-line document registering and retrieving system for AR annotation overlay. / Uchiyama, Hideaki; Pilet, Julien; Saito, Hideo.

ACM International Conference Proceeding Series. 2010. 1785478.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Uchiyama, H, Pilet, J & Saito, H 2010, On-line document registering and retrieving system for AR annotation overlay. in ACM International Conference Proceeding Series., 1785478, 1st Augmented Human International Conference, AH'10, Megeve, France, 10/4/2. https://doi.org/10.1145/1785455.1785478
Uchiyama H, Pilet J, Saito H. On-line document registering and retrieving system for AR annotation overlay. In ACM International Conference Proceeding Series. 2010. 1785478 https://doi.org/10.1145/1785455.1785478
Uchiyama, Hideaki ; Pilet, Julien ; Saito, Hideo. / On-line document registering and retrieving system for AR annotation overlay. ACM International Conference Proceeding Series. 2010.
@inproceedings{177c2d61477d404dbf2562aecaf67955,
title = "On-line document registering and retrieving system for AR annotation overlay",
abstract = "We propose a system that registers and retrieves text documents to annotate them on-line. The user registers a text document captured from a nearly top view and adds virtual annotations. When the user thereafter captures the document again, the system retrieves and displays the appropriate annotations, in real-time and at the correct location. Registering and deleting documents is done by user interaction. Our approach relies on LLAH, a hashing based method for document image retrieval. At the on-line registering stage, our system extracts keypoints from the input image and stores their descriptors computed from their neighbors. After registration, our system can quickly find the stored document corresponding to an input view by matching keypoints. From the matches, our system estimates the geometrical relationship between the camera and the document for accurately overlaying the annotations. In the experimental results, we show that our system can achieve on-line and real-time performances.",
keywords = "augmented reality, document retrieval, feature matching, LLAH, Poes estimation",
author = "Hideaki Uchiyama and Julien Pilet and Hideo Saito",
year = "2010",
doi = "10.1145/1785455.1785478",
language = "English",
isbn = "9781605588254",
booktitle = "ACM International Conference Proceeding Series",

}

TY - GEN

T1 - On-line document registering and retrieving system for AR annotation overlay

AU - Uchiyama, Hideaki

AU - Pilet, Julien

AU - Saito, Hideo

PY - 2010

Y1 - 2010

N2 - We propose a system that registers and retrieves text documents to annotate them on-line. The user registers a text document captured from a nearly top view and adds virtual annotations. When the user thereafter captures the document again, the system retrieves and displays the appropriate annotations, in real-time and at the correct location. Registering and deleting documents is done by user interaction. Our approach relies on LLAH, a hashing based method for document image retrieval. At the on-line registering stage, our system extracts keypoints from the input image and stores their descriptors computed from their neighbors. After registration, our system can quickly find the stored document corresponding to an input view by matching keypoints. From the matches, our system estimates the geometrical relationship between the camera and the document for accurately overlaying the annotations. In the experimental results, we show that our system can achieve on-line and real-time performances.

AB - We propose a system that registers and retrieves text documents to annotate them on-line. The user registers a text document captured from a nearly top view and adds virtual annotations. When the user thereafter captures the document again, the system retrieves and displays the appropriate annotations, in real-time and at the correct location. Registering and deleting documents is done by user interaction. Our approach relies on LLAH, a hashing based method for document image retrieval. At the on-line registering stage, our system extracts keypoints from the input image and stores their descriptors computed from their neighbors. After registration, our system can quickly find the stored document corresponding to an input view by matching keypoints. From the matches, our system estimates the geometrical relationship between the camera and the document for accurately overlaying the annotations. In the experimental results, we show that our system can achieve on-line and real-time performances.

KW - augmented reality

KW - document retrieval

KW - feature matching

KW - LLAH

KW - Poes estimation

UR - http://www.scopus.com/inward/record.url?scp=77954487840&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77954487840&partnerID=8YFLogxK

U2 - 10.1145/1785455.1785478

DO - 10.1145/1785455.1785478

M3 - Conference contribution

AN - SCOPUS:77954487840

SN - 9781605588254

BT - ACM International Conference Proceeding Series

ER -