生成对抗网络赋能面向平面布局生成中的家具布置  

Generative Adversarial Networks Empowered Furniture Arrangement for 2D Layout Configuration

作　　者：梁理锋 李光耀[1] LIANG Lifeng;LI Guangyao(School of Chemical and Materials Engineering,Zhejiang A&F University,Hangzhou 311300,China)

机构地区：[1]浙江农林大学化学与材料工程学院,杭州311300

出　　处：《家具》2025年第1期23-28,共6页Furniture

基　　金：浙江省自然科学基金(LQY18C160001)。

摘　　要：传统的家具平面布置图设计因其专业性、主观性以及设计流程耗时耗力,在寻找合适的辅助设计方面遇到挑战。生成式AI是实现设计智能化的有效辅助工具,通过将生成式AI融入家具平面布置图设计显示出潜力。鉴于此,研究采用生成对抗网络算法辅助设计师提升效率,通过分析生成对抗网络的发散性与交互性,以及家具平面布局设计任务自身多样化以及协同化的要求,指明了生成对抗网络在家具平面布置图设计任务中的应用适宜性,并结合建筑平面布局领域的先进研究提出了针对家具平面布置图设计任务的一些方法与建议,提出了基于功能泡泡图的家具布置图数据处理方式以及生成器与判别器架构的具体设置思路,最后讨论了评价体系,研究结果认为GAN运用于家具布置设计是一条有效的路径。Traditional furniture layout design faces challenges due to its specialized and subjective nature,as well as the labor-intensive and time-consuming design process.Generative AI has emerged as an effective tool to facilitate the intelligent design of furniture layouts,demonstrating significant potential.This study employs generative adversarial network(GANs)algorithms to enhance designers'efficiency.By examining the divergence and interactivity of GANs,along with the diverse and collaborative demands of furniture layout design tasks,the suitability of GANs for these tasks is highlighted.Furthermore,,the study synthesizes advanced research from the architectural layout field to propose methods and recommendations for furniture layout design tasks,including approaches for graph data processing,network configurations,and evaluation systems.This aims to provide a comprehensive reference for the intelligent and data-driven design of furniture layout plans.

关 键 词：生成式AI 生成对抗网络 家具平面布置图 辅助设计 

分 类 号：TS664[轻工技术与工程]