2008 IEEE Second International Conference on Biometrics: Theory, Applications and Systems 2008
DOI: 10.1109/btas.2008.4699374
|View full text |Cite
|
Sign up to set email alerts
|

Synthesizing Realistic Expressions in 3D Face Data Sets

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2

Citation Types

0
4
0

Year Published

2010
2010
2013
2013

Publication Types

Select...
5
1

Relationship

0
6

Authors

Journals

citations
Cited by 8 publications
(4 citation statements)
references
References 13 publications
0
4
0
Order By: Relevance
“…4) It is helpful for face synthesis. Most of the existing methods for face synthesis work with neutral faces [11,21,33,17]. Given an expressional face, changing it to another expression is beyond the capability of the current methods.…”
Section: Introductionmentioning
confidence: 99%
See 1 more Smart Citation
“…4) It is helpful for face synthesis. Most of the existing methods for face synthesis work with neutral faces [11,21,33,17]. Given an expressional face, changing it to another expression is beyond the capability of the current methods.…”
Section: Introductionmentioning
confidence: 99%
“…The current 3D expression synthesis approaches can be roughly classified into four categories: 1) Interpolation-based methods define an interpolation function to specify smooth motions between two key frames [18,21]. Although interpolation usually is fast, they cannot generate arbitrary realistic facial expressions ; 2)Approaches based on physical muscle models propagate muscle strength in an elastic spring mesh to model facial expressions [6,30,10]; 3) Example-based methods synthesize expressions from a neutral face based on a collections of examples [11,17]. It can synthesize expressions in real-time with low computational cost; 4) Pseudo muscle models simulate muscle forces by splines, tensors and free deformation models [27,33,24].…”
Section: Introductionmentioning
confidence: 99%
“…The only issue caused by the system is that it only covers the frontal face. They are focusing on making the transfer of expression for 3D sketches in the future.Minoi et.al [22]. developed a robust system that would transfer realistic expressions to a 3D face model captured by frontal photographs with a neutral face.…”
mentioning
confidence: 99%
“…The 3D face model created can render in a variety of poses and illumination. "The SDM approach extracts expression discriminant information efficiently, providing a gradual transformation on the 3D faces[22]". The main advantage of the system is that the variety of expressions can be easily generated without performing intense changes in the 3D database.…”
mentioning
confidence: 99%