Share Email Print

Proceedings Paper

Construction and exploitation of a 3D model from 2D image features
Author(s): Karl Ni; Zachary Sun; Nadya Bliss; Noah Snavely
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

This paper proposes a trainable computer vision approach for visual object registration relative to a collection of training images obtained a priori. The algorithm first identifies whether or not the image belongs to the scene location, and should it belong, it will identify objects of interest within the image and geo-register them. To accomplish this task, the processing chain relies on 3-D structure derived from motion to represent feature locations in a proposed model. Using current state-of- the-art algorithms, detected objects are extracted and their two-dimensional sizes in pixel quantities are converted into relative 3-D real-world coordinates using scene information, homography, and camera geometry. Locations can then be given with distance alignment information. The tasks can be accomplished in an efficient manner. Finally, algorithmic evaluation is presented with receiver operating characteristics, computational analysis, and registration errors in physical distances.

Paper Details

Date Published: 27 January 2010
PDF: 10 pages
Proc. SPIE 7533, Computational Imaging VIII, 75330J (27 January 2010); doi: 10.1117/12.849919
Show Author Affiliations
Karl Ni, MIT Lincoln Lab. (United States)
Zachary Sun, MIT Lincoln Lab. (United States)
Boston Univ. (United States)
Nadya Bliss, MIT Lincoln Lab. (United States)
Noah Snavely, Cornell Univ. (United States)

Published in SPIE Proceedings Vol. 7533:
Computational Imaging VIII
Charles A. Bouman; Ilya Pollak; Patrick J. Wolfe, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?