Effective Face Frontalization in Unconstrained Images

  Supriya Kamble,  Milind Rane,  Rutuja Vartale,  Kanchan Shinde

The Procedure of converting unconstrained (side viewed) facing images into front one are called Frontalization Process. Recent reports says that this method could well boost the performance of face recognition systems. This, by remodeling the difficult drawback of recognizing faces viewed from different viewpoints to the simpler drawback of recognizing faces in unnatural, frontal facing poses. Previous frontalization methods did this by making an attempt to approximate 3D facial shapes for every query image. We observe that 3D face shape estimation from unconstrained photos may be a harder problem than frontalization and can potentially introduce facial misalignments. Instead, we explore the simpler approach of using a single, unmodified, 3D surface as an approximation to the shape of all input faces. We show that this leads to a straightforward ,efficient and easy to implement method for frontalization. More significantly, it produces aesthetic new frontal views and effective method used for face recognition and gender estimation.

Keywords- Face Frontalization, 3D Facial Shapes, Gender Estimation.

Unique Identification Number - IJEDR2002077
Page Number(s) - 453-456
Pubished in - Volume 8 | Issue 2 | April 2020
DOI (Digital Object Identifier) -    http://doi.one/10.1729/Journal.23955
Publisher - IJEDR (ISSN - 2321-9939)

  Supriya Kamble,  Milind Rane,  Rutuja Vartale,  Kanchan Shinde,   "Effective Face Frontalization in Unconstrained Images", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.8, Issue 2, pp.453-456, April 2020, Available at :http://www.ijedr.org/papers/IJEDR2002077.pdf