TY - GEN
T1 - Focusing patch
T2 - 21st International Conference on MultiMedia Modeling, MMM 2015
AU - Kawai, Masahide
AU - Morishima, Shigeo
PY - 2015
Y1 - 2015
N2 - Facial image synthesis creates blurred facial images almost without high-frequency components, resulting in flat edges. Moreover, the synthesis process results in inconsistent facial images, such as the conditions where the white part of the eye is tinged with the color of the iris and the nasal cavity is tinged with the skin color. Therefore, we propose a method that can deblur an inconsistent synthesized facial image, including strong blurs created by common image morphing methods, and synthesize photographic quality facial images as clear as an image captured by a camera. Our system uses two original algorithms: patch color transfer and patch-optimized visio-lization. Patch color transfer can normalize facial luminance values with high precision, and patch-optimized visio-lization can synthesize a deblurred, photographic quality facial image. The advantages of our method are that it enables the reconstruction of the high-frequency components (concavo-convex) of human skin and removes strong blurs by employing only the input images used for original image morphing.
AB - Facial image synthesis creates blurred facial images almost without high-frequency components, resulting in flat edges. Moreover, the synthesis process results in inconsistent facial images, such as the conditions where the white part of the eye is tinged with the color of the iris and the nasal cavity is tinged with the skin color. Therefore, we propose a method that can deblur an inconsistent synthesized facial image, including strong blurs created by common image morphing methods, and synthesize photographic quality facial images as clear as an image captured by a camera. Our system uses two original algorithms: patch color transfer and patch-optimized visio-lization. Patch color transfer can normalize facial luminance values with high precision, and patch-optimized visio-lization can synthesize a deblurred, photographic quality facial image. The advantages of our method are that it enables the reconstruction of the high-frequency components (concavo-convex) of human skin and removes strong blurs by employing only the input images used for original image morphing.
KW - Blur
KW - Inconsistent facial image
KW - Patch color transfer
KW - Patchoptimized visio-lization
UR - http://www.scopus.com/inward/record.url?scp=84919629098&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84919629098&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-14445-0_14
DO - 10.1007/978-3-319-14445-0_14
M3 - Conference contribution
AN - SCOPUS:84919629098
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 155
EP - 166
BT - MultiMedia Modeling - 21st International Conference, MMM 2015, Proceedings
A2 - He, Xiangjian
A2 - Tao, Dacheng
A2 - Hasan, Muhammad Abul
A2 - Luo, Suhuai
A2 - Xu, Changsheng
A2 - Yang, Jie
PB - Springer Verlag
Y2 - 5 January 2015 through 7 January 2015
ER -