Skip to content

MohamedAmirr/graduation-project

Repository files navigation

Story Image Generator "IMAGITALE"

Project Overview:

IMAGITALE is designed to empower users to create and generate their own unique Disney-style stories and images. The system utilizes advanced AI models to transform written narratives into vivid images, providing a seamless and interactive storytelling experience.

Technologies Used:

  1. OpenAI GPT-3.5: This powerful language model is used to generate detailed and coherent stories from user inputs.
  2. Stable Diffusion XL (SDXL): A state-of-the-art image generation model that creates high-resolution, realistic images based on the story descriptions provided.
  3. DreamBooth and LoRA: Techniques used for fine-tuning and improving the consistency and quality of generated images.

Website Features:

  1. User-Friendly Interface: Easy navigation and intuitive design to ensure accessibility for users of all ages and technical backgrounds.
  2. Creative Empowerment: Tools for users to craft personalized stories with visual elements.
  3. Educational Value: Helps in developing storytelling skills and creativity, especially among children.
  4. Story and Image Generation: Allows users to input story descriptions and generate corresponding images, with options to add multiple scenes.

Challenges and Solutions:

One significant challenge we faced was ensuring consistency in character representation across different scenes. To address this:

  1. Fine-Tuning with DreamBooth and LoRA: Enhanced the model to specialize in rendering specific subjects consistently.
  2. Metadata Storage: Used metadata to maintain consistency in the appearance of characters and scenes.
  3. Batch Generation: Improved the efficiency and consistency of image generation by optimizing GPU memory usage.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •