**Face Recognition Using a Hybrid ResNet-ConvNet Neural Network**
Ali Miraftab, Paul Rad
*Open Cloud Institute, University of Texas at San Antonio, San Antonio, Texas, USA*
{Ali.Miraftab, Paul.Rad}@utsa.edu
**Dataset:** The image data can be found in [http://www.cs.cmu.edu/afs/][1]. This directory contains 20 subdirectories, one for each person, named by userid. Each of these directories contains several different face images of the same person.
**Outcome:** Applying resnet and convnet to identify the direction of the face
**Project Definition:** Using deep neural network models, ResNet and Convnet, classification of camera images of faces of various people in various poses is sudied. The dataset includes images of 20 different people, approximately 32 images per person, varying the person's expression (happy, sad, angry, neutral), the direction in which they are looking (left, right, straight ahead, up), and whether or not they were wearing sunglassess.
There is also variation in the background behind the person, the clothing worn by the person, and theposition of the person's face within the image. I total. 624 greyscale images were collected, each whithin a resolutoin of 120by128, with each image pixel described by a greyscale intensity value between 0 (black) and 255 (white).
A variety of target function can be learn from this image data. For example, given an image as input we could train a model to output the identity of the person, the direction in which the person is facing, the gender of the person, whether or not they are wearning sunglasses, etc. All of this target functions can be learned to high accuracy from this image data. In this course research we consider the particular task: learning the direction in which the person is facing (to their left, right, straight ahead, or upward)[1].
[1]: Mitchell, T. M. (1997). Machine Learning. New York: McGraw-Hill.