Exploring text-to-image generators for enhancing artistic creativity for cartoon aesthetics
Abstract
Recent advancements in the field of generative artificial intelligence have made it possible for text-to-image models to generate high quality images from natural language prompts. Despite these advancements, generating outputs such as cartoon aesthetics remains a challenge due to the need for structured prompts and generation parameters. This can limit users without any prior art experience or those who have not used such models before. This project therefore focuses on the development of an interactive image generation system which aims to allow users with no prior art experience to visualize their ideas through a simple and intuitive text input, with the output emphasizing on generating cartoon-style images. The system leverages on the Stable Diffusion 3.5 Large (SD3.5) model as its core model. The system is also built using a modular architecture which separated the inference pipeline from the user interface, allowing flexibility and scalability. The system supports multiple functions. These include text-to-image generation, image-to image generation, and inpainting for more targeted editing. A prompt helper powered by a large language model is also introduced. The prompt helper helps users to transform their inputs into more structured and model-friendly prompts. This improves output consistency and makes it more user friendly. The application is deployed using a Gradio interface. This allows users to interactively adjust generation parameters such as inference steps,…
Read the paper
Related studies
- Learning from First-Generation Qualitative Approaches in the IS Discipline: An Evolutionary View and Some Implications for Authors and Evaluators (PART 1/2)
- Commentary on the clinical and preclinical dosage limits of interstitially administered magnetic fluids for therapeutic hyperthermia based on current practice and efficacy models
- Prosthesis: New Technological Opportunities and Innovative Biomedical Devices
- The history and future of neural modeling for cochlear implants
- Ryanodine receptors: waking up from refractoriness
- Novel Transgenic Rabbit Model Sheds Light on the Puzzling Role of Matrix Metalloproteinase-12 in Atherosclerosis
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.