Publication Date

Spring 2023

Degree Type

Master's Project

Degree Name

Master of Science (MS)


Computer Science

First Advisor

Nada Attar

Second Advisor

Navrati Saxena

Third Advisor

Abhishek Chintala


Computer Vision, Facial Recognition, Image Classification, Bias, CNN


Computer Vision has been quickly transforming the way we live and work. One of its sub- domains, i.e., Facial Recognition has also been advancing at a rapid pace. However, the development of machine learning models that power these systems has been marred by social biases, which open the door to various societal issues. The objective of this project is to address these issues and ensure that computer vision systems are unbiased and fair to all individuals. To achieve this, we have created a web tool that uses three image classifiers (implemented using CNNs) to classify images into categories based on ethnicity, age, and gender. This tool is used to identify social biases present in datasets, such as if there are more women than men in a particular dataset, or if there are more children than senior citizens. The tool achieved a relatively small error rate of 5.00% for ethnicity, 3.75% for gender, and 8.33% for age. By identifying and dealing with social biases present in the datasets, we hope that facial recognition systems can be more equitable for all. Facial Recognition is being extensively used in many industries such as healthcare, retail, and security, making it imperative to address these issues. As advances in computer vision and facial recognition continue, it is crucial to ensure that these systems are unbiased and fair to all individuals, and the work done in this project sets a foundation to build upon in addressing this serious problem and its potentially far-reaching consequences.