Sasan Karamizadeh

A Hybrid Vision Transformer and EfficientNet-B3 Framework for Facial Expression Recognition

A hybrid architecture that combines Vision Transformers (ViTs) to capture global context with EfficientNet-B3 for multi-scale feature extraction and highlights the promise of hybrid deep learning architectures in tackling real-world facial expression recognition challenges.