Concept Generation
Concept generation, the process of creating novel ideas and designs, is increasingly being aided by artificial intelligence, particularly generative models. Current research focuses on leveraging large language models (LLMs) and diffusion models, often combined with Computer-Aided Design (CAD) data or iterative refinement techniques, to generate feasible and diverse design concepts across various domains, from engineering to automotive design. This work aims to improve the efficiency and creativity of the design process by automating aspects of idea generation and exploration, ultimately impacting both the speed and quality of innovation in numerous fields.
Papers
Idea2Img: Iterative Self-Refinement with GPT-4V(ision) for Automatic Image Design and Generation
Zhengyuan Yang, Jianfeng Wang, Linjie Li, Kevin Lin, Chung-Ching Lin, Zicheng Liu, Lijuan Wang
SingleInsert: Inserting New Concepts from a Single Image into Text-to-Image Models for Flexible Editing
Zijie Wu, Chaohui Yu, Zhen Zhu, Fan Wang, Xiang Bai