Ali Zoljodi and Hamid Mousavi: Advancements in Deep Learning for Computer Vision

Abstract: This seminar introduces Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs), both integral to advancing computer vision technology. We'll explore the underlying principles and functionalities of CNNs, renowned for their proficiency in image analysis, and ViTs, which adapt self-attention mechanisms for visual content. The session will highlight the distinct architectures of each, showcase their applications across various sectors, and discuss recent innovations that boost their performance in complex vision tasks.