Back to Projects

Facial Emotion Recognition

Jan 2024

A state-of-the-art computer vision system capable of accurately detecting and classifying human facial emotions.

Key Features

👁️

Vision Transformers (ViT)

Built a high-performance model using Vision Transformers (ViT), fine-tuned on a custom dataset to achieve superior accuracy compared to traditional CNNs.

🎯

Advanced Optimization

Implemented advanced optimization techniques including Mixup, Label Smoothing, and Test Time Augmentation (TTA) to improve model generalization and robustness.

Technologies

PythonPyTorchVision TransformersStreamlitDocker