Virtual Object Overlay onto Uncalibrated Camera Image Sequence Enabling Tracking of Natural Features
Author(s) -
Shunsuke Harasaki,
Hideo Saitô
Publication year - 2003
Publication title -
journal of robotics and mechatronics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.257
H-Index - 19
eISSN - 1883-8049
pISSN - 0915-3942
DOI - 10.20965/jrm.2003.p0263
Subject(s) - computer vision , artificial intelligence , computer science , overlay , tracking (education) , object (grammar) , augmented reality , virtual image , sequence (biology) , computer graphics (images) , image registration , video tracking , base (topology) , image (mathematics) , mathematics , psychology , mathematical analysis , pedagogy , biology , genetics , programming language
We discuss new geometric registration overlaying a virtual object onto an uncalibrated camera image sequence enabling extracting and tracking of natural features. To achieve correct registration, we apply a special space defined by two base images and trifocal tensors between base images and uncalibrated camera images. By extracting and tracking natural features, we construct augmented reality without markers whose 3D positions are known. Since we apply points and lines, fewer registration errors occur.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom