Application of Prompt Engineering in AIGC — Taking Stable Diffusion as an Example
扩散
计算机科学
热力学
物理
作者
Tingshuai Chen,Ye Yuan,Bingyang Yin
标识
DOI:10.1109/mlise62164.2024.10674412
摘要
In recent years, AIGC technology has gradually received attention from many fields. The emergence of AI painting generation tools such as Stable Diffusion has had a profound impact on the design field, especially in optimizing and simplifying the product design process. At present, prompt words are a major challenge. This article proposes a design framework for prompt word engineering, aiming to share how to quickly write prompt words that meet one's own needs based on product design concepts and creative ideas, and use Stable Diffusion to quickly achieve creative ideas and generate product design drawings. At the same time, a LoRA fine-tuning model is also trained to guide product style, and the ControlNet network is combined to control the generated content, thereby generating a series of products. Through the experience and reflection during the experimental process, we hope that this article can provide valuable reference and inspiration for peers.