DEEPFOCAL: A method for direct focal length estimation

Scott Workman; Connor Greenwell; Menghua Zhai; Ryan Baltenberger; Nathan Jacobs

doi:10.1109/ICIP.2015.7351024

DEEPFOCAL: A method for direct focal length estimation

Scott Workman, Connor Greenwell, Menghua Zhai, Ryan Baltenberger, Nathan Jacobs

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

71 Citations (SciVal)

Abstract

Estimating the focal length of an image is an important preprocessing step for many applications. Despite this, existing methods for single-view focal length estimation are limited in that they require particular geometric calibration objects, such as orthogonal vanishing points, co-planar circles, or a calibration grid, to occur in the field of view. In this work, we explore the application of a deep convolutional neural network, trained on natural images obtained from Internet photo collections, to directly estimate the focal length using only raw pixel intensities as input features. We present quantitative results that demonstrate the ability of our technique to estimate the focal length with comparisons against several baseline methods, including an automatic method which uses orthogonal vanishing points.

Original language	English
Title of host publication	2015 IEEE International Conference on Image Processing, ICIP 2015 - Proceedings
Pages	1369-1373
Number of pages	5
ISBN (Electronic)	9781479983391
DOIs	https://doi.org/10.1109/ICIP.2015.7351024
State	Published - Dec 9 2015
Event	IEEE International Conference on Image Processing, ICIP 2015 - Quebec City, Canada Duration: Sep 27 2015 → Sep 30 2015

Publication series

Name	Proceedings - International Conference on Image Processing, ICIP
Volume	2015-December
ISSN (Print)	1522-4880

Conference

Conference	IEEE International Conference on Image Processing, ICIP 2015
Country/Territory	Canada
City	Quebec City
Period	9/27/15 → 9/30/15

Bibliographical note

Publisher Copyright:
© 2015 IEEE.

Keywords

camera calibration
convolutional neural network
focal length estimation

ASJC Scopus subject areas

Software
Computer Vision and Pattern Recognition
Signal Processing

Access to Document

10.1109/ICIP.2015.7351024

Cite this

Workman, S., Greenwell, C., Zhai, M., Baltenberger, R., & Jacobs, N. (2015). DEEPFOCAL: A method for direct focal length estimation. In 2015 IEEE International Conference on Image Processing, ICIP 2015 - Proceedings (pp. 1369-1373). Article 7351024 (Proceedings - International Conference on Image Processing, ICIP; Vol. 2015-December). https://doi.org/10.1109/ICIP.2015.7351024

@inproceedings{09525ae48a2041789462b93743c9e0d3,

title = "DEEPFOCAL: A method for direct focal length estimation",

abstract = "Estimating the focal length of an image is an important preprocessing step for many applications. Despite this, existing methods for single-view focal length estimation are limited in that they require particular geometric calibration objects, such as orthogonal vanishing points, co-planar circles, or a calibration grid, to occur in the field of view. In this work, we explore the application of a deep convolutional neural network, trained on natural images obtained from Internet photo collections, to directly estimate the focal length using only raw pixel intensities as input features. We present quantitative results that demonstrate the ability of our technique to estimate the focal length with comparisons against several baseline methods, including an automatic method which uses orthogonal vanishing points.",

keywords = "camera calibration, convolutional neural network, focal length estimation",

author = "Scott Workman and Connor Greenwell and Menghua Zhai and Ryan Baltenberger and Nathan Jacobs",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; IEEE International Conference on Image Processing, ICIP 2015 ; Conference date: 27-09-2015 Through 30-09-2015",

year = "2015",

month = dec,

day = "9",

doi = "10.1109/ICIP.2015.7351024",

language = "English",

series = "Proceedings - International Conference on Image Processing, ICIP",

pages = "1369--1373",

booktitle = "2015 IEEE International Conference on Image Processing, ICIP 2015 - Proceedings",

}

Workman, S, Greenwell, C, Zhai, M, Baltenberger, R & Jacobs, N 2015, DEEPFOCAL: A method for direct focal length estimation. in 2015 IEEE International Conference on Image Processing, ICIP 2015 - Proceedings., 7351024, Proceedings - International Conference on Image Processing, ICIP, vol. 2015-December, pp. 1369-1373, IEEE International Conference on Image Processing, ICIP 2015, Quebec City, Canada, 9/27/15. https://doi.org/10.1109/ICIP.2015.7351024

DEEPFOCAL: A method for direct focal length estimation. / Workman, Scott; Greenwell, Connor; Zhai, Menghua et al.
2015 IEEE International Conference on Image Processing, ICIP 2015 - Proceedings. 2015. p. 1369-1373 7351024 (Proceedings - International Conference on Image Processing, ICIP; Vol. 2015-December).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - DEEPFOCAL

T2 - IEEE International Conference on Image Processing, ICIP 2015

AU - Workman, Scott

AU - Greenwell, Connor

AU - Zhai, Menghua

AU - Baltenberger, Ryan

AU - Jacobs, Nathan

PY - 2015/12/9

Y1 - 2015/12/9

N2 - Estimating the focal length of an image is an important preprocessing step for many applications. Despite this, existing methods for single-view focal length estimation are limited in that they require particular geometric calibration objects, such as orthogonal vanishing points, co-planar circles, or a calibration grid, to occur in the field of view. In this work, we explore the application of a deep convolutional neural network, trained on natural images obtained from Internet photo collections, to directly estimate the focal length using only raw pixel intensities as input features. We present quantitative results that demonstrate the ability of our technique to estimate the focal length with comparisons against several baseline methods, including an automatic method which uses orthogonal vanishing points.

AB - Estimating the focal length of an image is an important preprocessing step for many applications. Despite this, existing methods for single-view focal length estimation are limited in that they require particular geometric calibration objects, such as orthogonal vanishing points, co-planar circles, or a calibration grid, to occur in the field of view. In this work, we explore the application of a deep convolutional neural network, trained on natural images obtained from Internet photo collections, to directly estimate the focal length using only raw pixel intensities as input features. We present quantitative results that demonstrate the ability of our technique to estimate the focal length with comparisons against several baseline methods, including an automatic method which uses orthogonal vanishing points.

KW - camera calibration

KW - convolutional neural network

KW - focal length estimation

UR - http://www.scopus.com/inward/record.url?scp=84956683598&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84956683598&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2015.7351024

DO - 10.1109/ICIP.2015.7351024

M3 - Conference contribution

AN - SCOPUS:84956683598

T3 - Proceedings - International Conference on Image Processing, ICIP

SP - 1369

EP - 1373

BT - 2015 IEEE International Conference on Image Processing, ICIP 2015 - Proceedings

Y2 - 27 September 2015 through 30 September 2015

ER -

DEEPFOCAL: A method for direct focal length estimation

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this