GPU Based Face Recognition System for Authentication

  Meghna Mandloi,  Bhumika Agrawal,  Chelsi Gupta,  Divya Dwivedi,  Jayesh Surana

Face has significant role in identifying a person for authentication purpose in public places such as airport security. Face recognition has many real world applications including surveillance and authentication. Due to complex and multidimensional structure of face it requires huge computations therefore fast face recognition is required. One of the most successful template based techniques for face recognition is Principal Component Analysis (PCA) which is generally known as Eigen face approach. It suffers from the disadvantage of higher computation cost, despite its better recognition rate. With the increase in number of images in training database and also the resolution of images, the computational cost also increases. Graphics Processing Unit (GPU) is the solution for fast and efficient computation. GPUs have massively parallel multi-threaded environment. With the use of GPU's parallel environment, a problem can be solved in parallel with much less time. NVIDIA has released a parallel programming framework CUDA (Compute Unified Device Architecture), which supports popular programming languages with CUDA extension for GPU programming. A parallel version of Eigen face approach for face recognition is developed using CUDA framework.

Keywords- GPU, CUDA, NVIDIA, PCA Eigenfaces, face recognition

Unique Identification Number - IJEDR1702155
Page Number(s) - 931-935
Pubished in - Volume 5 | Issue 2 | May 2017
DOI (Digital Object Identifier) -   
Publisher - IJEDR (ISSN - 2321-9939)

  Meghna Mandloi,  Bhumika Agrawal,  Chelsi Gupta,  Divya Dwivedi,  Jayesh Surana,   "GPU Based Face Recognition System for Authentication", International Journal of Engineering Development and Research (IJEDR), ISSN:2321-9939, Volume.5, Issue 2, pp.931-935, May 2017, Available at :http://www.ijedr.org/papers/IJEDR1702155.pdf