Abstract

Computational creativity refers to the study and development of algorithms, systems, and models that can exhibit creativity that resembles or complements human creative processes. With the exponential growth of digital content and multimedia data, automated multimedia content generation has become a pivotal research area in artificial intelligence (AI) and machine learning (ML). This paper investigates computational creativity models specifically in the context of generating multimedia content such as images, videos, music, and interactive narratives. We explore foundational theories of creativity, the adaptation of generative models—such as generative adversarial networks (GANs), variational autoencoders (VAEs), transformer architectures, and evolutionary algorithms—and their applications in multimedia creativity tasks. Through an extensive literature review, we examine developments from early rule-based and knowledge-based systems to state-of-the-art deep learning techniques. The research methodology outlines a framework for evaluating these models based on creativity metrics, quality assessments, and user perceptions. The study also discusses advantages and disadvantages, highlighting scalability, novelty, coherence, and ethical considerations. Results illustrate both the potential and limitations of current models, while conclusions and future work propose avenues for enhancing automated creativity in multimedia generation.