The integration of Artificial Intelligence (AI) into artistic domains has grown significantly in recent years, particularly through the advancements in Computer Vision (CV). CV is a field of AI focused on enabling machines to interpret and understand visual information trough automatic methods. In this work, we present a comprehensive overview of CV techniques applied to image analysis and generation. We begin by tracing the evolution from classical approaches of feature extraction and colour processing, to contemporary deep learning methods, such as convolutional neural networks (CNNs) and Transformers, achieving high-level semantic understanding of visual content. The study then shifts to image generation, examining different paradigms and exploring the various architectures involved in the task. Additionally, we explore multimodal models that integrate textual and visual data, highlighting their role in the conditional image generation. Finally, the paper touches on the ethical considerations surrounding AI-generated imagery, including concerns about authenticity, bias, and misuse. This work aims to provide a cohesive understanding of how AI is shaping the visual dimension of artistic and computational practices.