An Efficiency-Optimized Framework for Sequential Image Generation with Stable Diffusion Models

Diffusion models are a type of Generative Artificial Intelligence models that create data by adding noise to the data and gradually removing the noise to generate synthetic data. This research work focuses on visual content generation based on text input. We are using a computationally efficient variant of diffusion models namely, a stable diffusion model to create images from text. The novelty of this research work focuses on sequential image generation from text prompts. This approach has been successfully implemented to create visual storyboards, and diverse, contextually coherent images from text prompts. Hyperparameters have been fine-tuned to generate high-quality and context-aware images from text prompts.