Classifying Eyewear on GAN-Generated Faces Using Convolutional Neural Networks
In today's world of computer vision, Convolutional Neural Networks (CNNs) play a crucial role in analyzing and processing image data. This study explores how CNNs work alongside image data, specifically focusing on classifying eyewear on GAN-generated faces. By harnessing CNNs' ability to extract hierarchical features from images and the capability of Generative Adversarial Networks (GANs) to create realistic synthetic data, this research highlights advancements in image classification tasks. The study evaluates the performance of various machine learning algorithms in classifying eyewear on GAN-generated faces. Results show that Ridge regression, Lasso regression, and Support Vector Machines perform well on tabular data, with Ridge regression achieving the highest accuracy. Shifting to image data, CNNs outshine feedforward neural networks (FFNN), especially those with multiple convolutional layers. This underscores the importance of leveraging CNN architectures for accurate image classification tasks, with potential applications in e-commerce and virtual try-on systems.
Authors
- Poorna Sandamini Senaratne (ORCID: https://orcid.org/0009-0009-2008-2836)
- Shahd Alnofaie
Institutions
- University of Central Florida (US)
Publication Details
- Journal
- STARS (University of Central Florida)
- Published
- 2026-11-20
- Primary Topic
- Advanced Neural Network Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00