Crafting Effective Prompts: A Guideline for Successful Image Generation
Noor Wahyuni Ahmad, Suzana Ruslan · 2024
AI’s ability to convert text into high-quality visuals is revolutionizing a number of study sectors, including the creative industries. This work focuses on how prompt construction plays a crucial role in maximizing large language models’ (LLMs’) efficacy in image production. While prompt engineering encompasses many study fields, this work focuses primarily on how prompts should be designed to enhance the quality of generated images utilizing prompt anatomy as input. In this study, two student groups were compared: one group produced photos on their own without any supervision, and the other group was given tips on how to make efficient suggestions. Clarity, detail, relevance, aesthetic appeal, and inventiveness were the criteria used to assess the prompts and the final photos. According to the panel of experts evaluating the results, the group who received guided prompt creation generated photos that were both much better and more in line with the intended concept. This work emphasizes how crucial good prompt design is for AI-driven image production and shows that appropriate prompt crafting training can significantly improve output quality. These results offer a foundation for the more efficient application of AI technologies and have important ramifications for AI-assisted design, education, and the creative industries.