This course explores machine learning methodologies for image and video synthesis. Students will examine a range of techniques, from classical methods to advanced deep learning models, including Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), autoregressive models, and diffusion models. The course emphasizes practical applications, enabling students to develop tools for artistic expression and realistic image generation. Additionally, it covers image and video forensics methods for detecting synthetic content.