Wrinkles individuality representing aging simulation

Pavel A. Savkin, Daiki Kuwahara, Masahide Kawai, Takuya Kato, Shigeo Morishima

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

    2 Citations (Scopus)

    Abstract

    An appearance of a human face changes due to aging: sagging, spots, lusters, and wrinkles would be observed. Therefore, facial aging simulation techniques are required for long-term criminal investigation. While the appearance of an aged face varies greatly from person to person, wrinkles are one of the most important features which represent the human individuality. An individuali-ty of wrinkles is defined by wrinkles shape and position. [Maejima et al. 2014] proposed an aging simulation method that preserves facial parts individuality using a patch-based facial image reconstruction. Since few wrinkles can be observed on an input young face, it is difficult to represent wrinkles appearance only by the reconstruction. Therefore, a statistical wrinkles aging pattern model is introduced to produce natural looking wrinkles by selecting appropriate patches in an age-specific patch database. However, the variation of the statistical wrinkles patterns model is too limited to represent wrinkles individuality. Additionally, an appropriate size of a patch and feature value had to be applied for each facial region to get a plausible aged facial image. In this paper, we introduce a novel aging simulation method us-ing the patch-based image reconstruction, which can overcome problems mentioned above. Based on a medical knowledge [Piér-ard et al. 2003], wrinkles in an expressive facial image (defined as expressive wrinkles) of a same person are synthesized to the input image instead of the statistical wrinkles pattern model to represent wrinkles individuality. Furthermore, different sizes of the patch and feature values are applied for each facial region to achieve high representation of both the wrinkles individuality and the age-specific features. The entire process is performed automatically, and a plausible aged facial image is generated.

    Original languageEnglish
    Title of host publicationSIGGRAPH Asia 2015 Posters, SA 2015
    PublisherAssociation for Computing Machinery, Inc
    ISBN (Print)9781450339261
    DOIs
    Publication statusPublished - 2015 Nov 2
    EventSIGGRAPH Asia, SA 2015 - Kobe, Japan
    Duration: 2015 Nov 22015 Nov 6

    Other

    OtherSIGGRAPH Asia, SA 2015
    CountryJapan
    CityKobe
    Period15/11/215/11/6

    Fingerprint

    Aging of materials
    Image reconstruction

    ASJC Scopus subject areas

    • Human-Computer Interaction
    • Computer Graphics and Computer-Aided Design
    • Computer Vision and Pattern Recognition

    Cite this

    Savkin, P. A., Kuwahara, D., Kawai, M., Kato, T., & Morishima, S. (2015). Wrinkles individuality representing aging simulation. In SIGGRAPH Asia 2015 Posters, SA 2015 [a37] Association for Computing Machinery, Inc. https://doi.org/10.1145/2820926.2820942

    Wrinkles individuality representing aging simulation. / Savkin, Pavel A.; Kuwahara, Daiki; Kawai, Masahide; Kato, Takuya; Morishima, Shigeo.

    SIGGRAPH Asia 2015 Posters, SA 2015. Association for Computing Machinery, Inc, 2015. a37.

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

    Savkin, PA, Kuwahara, D, Kawai, M, Kato, T & Morishima, S 2015, Wrinkles individuality representing aging simulation. in SIGGRAPH Asia 2015 Posters, SA 2015., a37, Association for Computing Machinery, Inc, SIGGRAPH Asia, SA 2015, Kobe, Japan, 15/11/2. https://doi.org/10.1145/2820926.2820942
    Savkin PA, Kuwahara D, Kawai M, Kato T, Morishima S. Wrinkles individuality representing aging simulation. In SIGGRAPH Asia 2015 Posters, SA 2015. Association for Computing Machinery, Inc. 2015. a37 https://doi.org/10.1145/2820926.2820942
    Savkin, Pavel A. ; Kuwahara, Daiki ; Kawai, Masahide ; Kato, Takuya ; Morishima, Shigeo. / Wrinkles individuality representing aging simulation. SIGGRAPH Asia 2015 Posters, SA 2015. Association for Computing Machinery, Inc, 2015.
    @inproceedings{b0c62f67d9544c8f88b7c2dec9741a0d,
    title = "Wrinkles individuality representing aging simulation",
    abstract = "An appearance of a human face changes due to aging: sagging, spots, lusters, and wrinkles would be observed. Therefore, facial aging simulation techniques are required for long-term criminal investigation. While the appearance of an aged face varies greatly from person to person, wrinkles are one of the most important features which represent the human individuality. An individuali-ty of wrinkles is defined by wrinkles shape and position. [Maejima et al. 2014] proposed an aging simulation method that preserves facial parts individuality using a patch-based facial image reconstruction. Since few wrinkles can be observed on an input young face, it is difficult to represent wrinkles appearance only by the reconstruction. Therefore, a statistical wrinkles aging pattern model is introduced to produce natural looking wrinkles by selecting appropriate patches in an age-specific patch database. However, the variation of the statistical wrinkles patterns model is too limited to represent wrinkles individuality. Additionally, an appropriate size of a patch and feature value had to be applied for each facial region to get a plausible aged facial image. In this paper, we introduce a novel aging simulation method us-ing the patch-based image reconstruction, which can overcome problems mentioned above. Based on a medical knowledge [Pi{\'e}r-ard et al. 2003], wrinkles in an expressive facial image (defined as expressive wrinkles) of a same person are synthesized to the input image instead of the statistical wrinkles pattern model to represent wrinkles individuality. Furthermore, different sizes of the patch and feature values are applied for each facial region to achieve high representation of both the wrinkles individuality and the age-specific features. The entire process is performed automatically, and a plausible aged facial image is generated.",
    author = "Savkin, {Pavel A.} and Daiki Kuwahara and Masahide Kawai and Takuya Kato and Shigeo Morishima",
    year = "2015",
    month = "11",
    day = "2",
    doi = "10.1145/2820926.2820942",
    language = "English",
    isbn = "9781450339261",
    booktitle = "SIGGRAPH Asia 2015 Posters, SA 2015",
    publisher = "Association for Computing Machinery, Inc",

    }

    TY - GEN

    T1 - Wrinkles individuality representing aging simulation

    AU - Savkin, Pavel A.

    AU - Kuwahara, Daiki

    AU - Kawai, Masahide

    AU - Kato, Takuya

    AU - Morishima, Shigeo

    PY - 2015/11/2

    Y1 - 2015/11/2

    N2 - An appearance of a human face changes due to aging: sagging, spots, lusters, and wrinkles would be observed. Therefore, facial aging simulation techniques are required for long-term criminal investigation. While the appearance of an aged face varies greatly from person to person, wrinkles are one of the most important features which represent the human individuality. An individuali-ty of wrinkles is defined by wrinkles shape and position. [Maejima et al. 2014] proposed an aging simulation method that preserves facial parts individuality using a patch-based facial image reconstruction. Since few wrinkles can be observed on an input young face, it is difficult to represent wrinkles appearance only by the reconstruction. Therefore, a statistical wrinkles aging pattern model is introduced to produce natural looking wrinkles by selecting appropriate patches in an age-specific patch database. However, the variation of the statistical wrinkles patterns model is too limited to represent wrinkles individuality. Additionally, an appropriate size of a patch and feature value had to be applied for each facial region to get a plausible aged facial image. In this paper, we introduce a novel aging simulation method us-ing the patch-based image reconstruction, which can overcome problems mentioned above. Based on a medical knowledge [Piér-ard et al. 2003], wrinkles in an expressive facial image (defined as expressive wrinkles) of a same person are synthesized to the input image instead of the statistical wrinkles pattern model to represent wrinkles individuality. Furthermore, different sizes of the patch and feature values are applied for each facial region to achieve high representation of both the wrinkles individuality and the age-specific features. The entire process is performed automatically, and a plausible aged facial image is generated.

    AB - An appearance of a human face changes due to aging: sagging, spots, lusters, and wrinkles would be observed. Therefore, facial aging simulation techniques are required for long-term criminal investigation. While the appearance of an aged face varies greatly from person to person, wrinkles are one of the most important features which represent the human individuality. An individuali-ty of wrinkles is defined by wrinkles shape and position. [Maejima et al. 2014] proposed an aging simulation method that preserves facial parts individuality using a patch-based facial image reconstruction. Since few wrinkles can be observed on an input young face, it is difficult to represent wrinkles appearance only by the reconstruction. Therefore, a statistical wrinkles aging pattern model is introduced to produce natural looking wrinkles by selecting appropriate patches in an age-specific patch database. However, the variation of the statistical wrinkles patterns model is too limited to represent wrinkles individuality. Additionally, an appropriate size of a patch and feature value had to be applied for each facial region to get a plausible aged facial image. In this paper, we introduce a novel aging simulation method us-ing the patch-based image reconstruction, which can overcome problems mentioned above. Based on a medical knowledge [Piér-ard et al. 2003], wrinkles in an expressive facial image (defined as expressive wrinkles) of a same person are synthesized to the input image instead of the statistical wrinkles pattern model to represent wrinkles individuality. Furthermore, different sizes of the patch and feature values are applied for each facial region to achieve high representation of both the wrinkles individuality and the age-specific features. The entire process is performed automatically, and a plausible aged facial image is generated.

    UR - http://www.scopus.com/inward/record.url?scp=84959419615&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84959419615&partnerID=8YFLogxK

    U2 - 10.1145/2820926.2820942

    DO - 10.1145/2820926.2820942

    M3 - Conference contribution

    AN - SCOPUS:84959419615

    SN - 9781450339261

    BT - SIGGRAPH Asia 2015 Posters, SA 2015

    PB - Association for Computing Machinery, Inc

    ER -