Bayesian Tensor Approach for 3-D Face Modeling

Effectively modeling a collection of three-dimensional (3-D) faces is an important task in various applications, especially facial expression-driven ones, e.g., expression generation, retargeting, and synthesis. These 3-D faces naturally form a set of second-order tensors-one modality for identity a...

Full description

Saved in:
Bibliographic Details
Main Authors: TAO, Dacheng, SONG, Mingli, LI, Xuelong, SHEN, Jialie, SUN, Jimeng, WU, Xindong, Faloutsos, Christos, Maybank, Stephen J.
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2008
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/771
https://ink.library.smu.edu.sg/context/sis_research/article/1770/viewcontent/BayesianTensorApproach3_DFace_2008.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
id sg-smu-ink.sis_research-1770
record_format dspace
spelling sg-smu-ink.sis_research-17702017-03-23T02:42:13Z Bayesian Tensor Approach for 3-D Face Modeling TAO, Dacheng SONG, Mingli LI, Xuelong SHEN, Jialie SUN, Jimeng WU, Xindong Faloutsos, Christos Maybank, Stephen J. Effectively modeling a collection of three-dimensional (3-D) faces is an important task in various applications, especially facial expression-driven ones, e.g., expression generation, retargeting, and synthesis. These 3-D faces naturally form a set of second-order tensors-one modality for identity and the other for expression. The number of these second-order tensors is three times of that of the vertices for 3-D face modeling. As for algorithms, Bayesian data modeling, which is a natural data analysis tool, has been widely applied with great success; however, it works only for vector data. Therefore, there is a gap between tensor-based representation and vector-based data analysis tools. Aiming at bridging this gap and generalizing conventional statistical tools over tensors, this paper proposes a decoupled probabilistic algorithm, which is named Bayesian tensor analysis (BTA). Theoretically, BTA can automatically and suitably determine dimensionality for different modalities of tensor data. With BTA, a collection of 3-D faces can be well modeled. Empirical studies on expression retargeting also justify the advantages of BTA. 2008-10-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/771 info:doi/10.1109/TCSVT.2008.2002825 https://ink.library.smu.edu.sg/context/sis_research/article/1770/viewcontent/BayesianTensorApproach3_DFace_2008.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University 3-D face Bayesian inference Bayesian tensor analysis face expression synthesis face recognition Databases and Information Systems
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic 3-D face
Bayesian inference
Bayesian tensor analysis
face expression synthesis
face recognition
Databases and Information Systems
spellingShingle 3-D face
Bayesian inference
Bayesian tensor analysis
face expression synthesis
face recognition
Databases and Information Systems
TAO, Dacheng
SONG, Mingli
LI, Xuelong
SHEN, Jialie
SUN, Jimeng
WU, Xindong
Faloutsos, Christos
Maybank, Stephen J.
Bayesian Tensor Approach for 3-D Face Modeling
description Effectively modeling a collection of three-dimensional (3-D) faces is an important task in various applications, especially facial expression-driven ones, e.g., expression generation, retargeting, and synthesis. These 3-D faces naturally form a set of second-order tensors-one modality for identity and the other for expression. The number of these second-order tensors is three times of that of the vertices for 3-D face modeling. As for algorithms, Bayesian data modeling, which is a natural data analysis tool, has been widely applied with great success; however, it works only for vector data. Therefore, there is a gap between tensor-based representation and vector-based data analysis tools. Aiming at bridging this gap and generalizing conventional statistical tools over tensors, this paper proposes a decoupled probabilistic algorithm, which is named Bayesian tensor analysis (BTA). Theoretically, BTA can automatically and suitably determine dimensionality for different modalities of tensor data. With BTA, a collection of 3-D faces can be well modeled. Empirical studies on expression retargeting also justify the advantages of BTA.
format text
author TAO, Dacheng
SONG, Mingli
LI, Xuelong
SHEN, Jialie
SUN, Jimeng
WU, Xindong
Faloutsos, Christos
Maybank, Stephen J.
author_facet TAO, Dacheng
SONG, Mingli
LI, Xuelong
SHEN, Jialie
SUN, Jimeng
WU, Xindong
Faloutsos, Christos
Maybank, Stephen J.
author_sort TAO, Dacheng
title Bayesian Tensor Approach for 3-D Face Modeling
title_short Bayesian Tensor Approach for 3-D Face Modeling
title_full Bayesian Tensor Approach for 3-D Face Modeling
title_fullStr Bayesian Tensor Approach for 3-D Face Modeling
title_full_unstemmed Bayesian Tensor Approach for 3-D Face Modeling
title_sort bayesian tensor approach for 3-d face modeling
publisher Institutional Knowledge at Singapore Management University
publishDate 2008
url https://ink.library.smu.edu.sg/sis_research/771
https://ink.library.smu.edu.sg/context/sis_research/article/1770/viewcontent/BayesianTensorApproach3_DFace_2008.pdf
_version_ 1770570707064848384