Direct bundle estimation for recovery of shape, reflectance property and light position

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

5 Citations (Scopus)

Abstract

Given a set of images captured with a fixed camera while a point light source moves around an object, we can estimate the shape, reflectance property and texture of the object, as well as the positions of the light source. Our formulation is a large-scale nonlinear optimization that allows us to adjust the parameters so that the images synthesized from all of the parameters optimally fit the input images. This type of optimization, which is a variation of the bundle adjustment for structure and motion reconstruction, is often employed to refine a carefully constructed initial estimation. However, the initialization task often requires a great deal of labor, several special devices, or both. In the present paper, we describe (i) an easy method of initialization that does not require any special devices or a precise calibration and (ii) an efficient algorithm for the optimization. The efficiency of the optimization method enables us to use a simple initialization. For a set of synthesized images, the proposed method decreases the residual to zero. In addition, we show that various real objects, including toy models and human faces, can be successfully recovered.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages412-425
Number of pages14
Volume5304 LNCS
EditionPART 3
DOIs
Publication statusPublished - 2008
Event10th European Conference on Computer Vision, ECCV 2008 - Marseille, France
Duration: Oct 12 2008Oct 18 2008

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 3
Volume5304 LNCS
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other10th European Conference on Computer Vision, ECCV 2008
CountryFrance
CityMarseille
Period10/12/0810/18/08

Fingerprint

Reflectance
Bundle
Recovery
Initialization
Light sources
Large-scale Optimization
Optimization
Nonlinear Optimization
Optimization Methods
Texture
Adjustment
Calibration
Efficient Algorithms
Textures
Camera
Cameras
Face
Personnel
Decrease
Motion

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Migita, T., Ogino, S., & Shakunaga, T. (2008). Direct bundle estimation for recovery of shape, reflectance property and light position. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (PART 3 ed., Vol. 5304 LNCS, pp. 412-425). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5304 LNCS, No. PART 3). https://doi.org/10.1007/978-3-540-88690-7-31

Direct bundle estimation for recovery of shape, reflectance property and light position. / Migita, Tsuyoshi; Ogino, Shinsuke; Shakunaga, Takeshi.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 5304 LNCS PART 3. ed. 2008. p. 412-425 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5304 LNCS, No. PART 3).

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

Migita, T, Ogino, S & Shakunaga, T 2008, Direct bundle estimation for recovery of shape, reflectance property and light position. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 3 edn, vol. 5304 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 3, vol. 5304 LNCS, pp. 412-425, 10th European Conference on Computer Vision, ECCV 2008, Marseille, France, 10/12/08. https://doi.org/10.1007/978-3-540-88690-7-31
Migita T, Ogino S, Shakunaga T. Direct bundle estimation for recovery of shape, reflectance property and light position. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). PART 3 ed. Vol. 5304 LNCS. 2008. p. 412-425. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 3). https://doi.org/10.1007/978-3-540-88690-7-31
Migita, Tsuyoshi ; Ogino, Shinsuke ; Shakunaga, Takeshi. / Direct bundle estimation for recovery of shape, reflectance property and light position. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 5304 LNCS PART 3. ed. 2008. pp. 412-425 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 3).
@inproceedings{f90933d7dc084e678d2090c8be5eea8a,
title = "Direct bundle estimation for recovery of shape, reflectance property and light position",
abstract = "Given a set of images captured with a fixed camera while a point light source moves around an object, we can estimate the shape, reflectance property and texture of the object, as well as the positions of the light source. Our formulation is a large-scale nonlinear optimization that allows us to adjust the parameters so that the images synthesized from all of the parameters optimally fit the input images. This type of optimization, which is a variation of the bundle adjustment for structure and motion reconstruction, is often employed to refine a carefully constructed initial estimation. However, the initialization task often requires a great deal of labor, several special devices, or both. In the present paper, we describe (i) an easy method of initialization that does not require any special devices or a precise calibration and (ii) an efficient algorithm for the optimization. The efficiency of the optimization method enables us to use a simple initialization. For a set of synthesized images, the proposed method decreases the residual to zero. In addition, we show that various real objects, including toy models and human faces, can be successfully recovered.",
author = "Tsuyoshi Migita and Shinsuke Ogino and Takeshi Shakunaga",
year = "2008",
doi = "10.1007/978-3-540-88690-7-31",
language = "English",
isbn = "3540886893",
volume = "5304 LNCS",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
number = "PART 3",
pages = "412--425",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
edition = "PART 3",

}

TY - GEN

T1 - Direct bundle estimation for recovery of shape, reflectance property and light position

AU - Migita, Tsuyoshi

AU - Ogino, Shinsuke

AU - Shakunaga, Takeshi

PY - 2008

Y1 - 2008

N2 - Given a set of images captured with a fixed camera while a point light source moves around an object, we can estimate the shape, reflectance property and texture of the object, as well as the positions of the light source. Our formulation is a large-scale nonlinear optimization that allows us to adjust the parameters so that the images synthesized from all of the parameters optimally fit the input images. This type of optimization, which is a variation of the bundle adjustment for structure and motion reconstruction, is often employed to refine a carefully constructed initial estimation. However, the initialization task often requires a great deal of labor, several special devices, or both. In the present paper, we describe (i) an easy method of initialization that does not require any special devices or a precise calibration and (ii) an efficient algorithm for the optimization. The efficiency of the optimization method enables us to use a simple initialization. For a set of synthesized images, the proposed method decreases the residual to zero. In addition, we show that various real objects, including toy models and human faces, can be successfully recovered.

AB - Given a set of images captured with a fixed camera while a point light source moves around an object, we can estimate the shape, reflectance property and texture of the object, as well as the positions of the light source. Our formulation is a large-scale nonlinear optimization that allows us to adjust the parameters so that the images synthesized from all of the parameters optimally fit the input images. This type of optimization, which is a variation of the bundle adjustment for structure and motion reconstruction, is often employed to refine a carefully constructed initial estimation. However, the initialization task often requires a great deal of labor, several special devices, or both. In the present paper, we describe (i) an easy method of initialization that does not require any special devices or a precise calibration and (ii) an efficient algorithm for the optimization. The efficiency of the optimization method enables us to use a simple initialization. For a set of synthesized images, the proposed method decreases the residual to zero. In addition, we show that various real objects, including toy models and human faces, can be successfully recovered.

UR - http://www.scopus.com/inward/record.url?scp=57149145413&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=57149145413&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-88690-7-31

DO - 10.1007/978-3-540-88690-7-31

M3 - Conference contribution

SN - 3540886893

SN - 9783540886891

VL - 5304 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 412

EP - 425

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -