School of Electrical and Computer Engineering, University of Tehran
I am pleased to offer the course “Deep Learning for Computer Vision” in Spring 2026. The course is designed to provide an application-oriented perspective on modern computer vision, aligning its content with leading international curricula.
Topics include modern vision architectures, pre-training and transfer learning, image classification, detection and segmentation, metric learning and retrieval, robustness and anomaly detection, foundation and vision-language models, video understanding and tracking, super-resolution, and 3D vision.
Unlike foundational Deep Learning courses (focused on core methodologies), classical Computer Vision courses (centered on traditional vision techniques), and Generative Modeling courses (primarily devoted to data generation), this course emphasizes modern deep-learning–driven
computer vision applications, with a focus on problem formulation, evolving solution paradigms, and the training and evaluation of state-of-the-art models in practical settings.
Evaluation will be based on projects, assignments, as well as a midterm and final exam, with emphasis on reading and implementing state-of-the-art research.
