学术论文|基于机器学习的三维空间学习方法的探索--以太湖石为例
基于机器学习的三维空间学习方法的探索——以太湖石为例
Exploration of three-dimensional spatial learning approach based on machine learning--taking Taihu stone as an example
邓巧明,李晓峰,刘宇波,胡凯 | Qiaoming Deng (Corresponding Author), Xiaofeng Li, Yubo Liu , Kai Hu
原文发表于Architectural Intelligence,未经允许,不得转载。
引用格式:
Deng, Q., Li, X., Liu, Y. et al. Exploration of three-dimensional spatial learning approach based on machine learning–taking Taihu stone as an example. ARIN 2, 5 (2023). https://doi.org/10.1007/s44223-023-00023-2
作者信息:
邓巧明 华南理工大学建筑学院副教授、硕士生导师,亚热带建筑科学国家重点实验室
李晓峰 华南理工大学建筑学院硕士研究生
刘宇波 华南理工大学建筑学院建筑系主任、教授、博士生导师,亚热带建筑科学国家重点实验室
胡 凯 华南理工大学建筑学院博士研究生
基金项目:
国家自然科学基金资助项目(51978268,51978269)
亚热带建筑科学国家重点实验室国际合作项目(2019ZA01)
论文摘要
关键词:机器学习,人工神经网络,Pix2Pix,空间转换,太湖石
Under the influence of globalization, the transformation of traditional architectural space is vital to the growth of local architecture. As an important spatial element of traditional gardens, Taihu stone has the image qualities of being "thin, wrinkled, leaky and transparent" The "transparency" and " leaky" of Taihu stone reflect the connectivity and irregularity of Taihu stone's holes, which are consistent with the contemporary architectural design concepts of fluid space and transparency. Nonetheless, relatively few theoretical studies have been conducted on the spatial analysis and design transformation of Taihu stone. Using machine learning, we attempt to extract the three-dimensional spatial variation pattern of Taihu stone in this paper. This study extracts 3D spatial features for experiments using artificial neural networks (ANN) and generative adversarial networks (GAN). In order to extract 3D spatial variation patterns, the machine learning model learns the variation patterns between adjacent sections. The trained machine learning model is capable of generating a series of spatial sections with the spatial variation pattern of the Taihu stone. The purpose of the experimental results is to compare the performance of various machine learning models for 3D space learning in order to identify a model with superior performance. This paper also presents a novel concept for machine learning to master continuous 3D spatial features.
Keywords: Machine learning · Artificial neural networks · Pix2Pix · Spatial transformation · Taihu stone.
主要图表
图1:研究流程
图2:太湖石原始数据样本
图4: ANN的控制点数据集
图5:ANN实验超参数优化过程
图6:ANN生成的三维空间模型
图7:基于空间分析的样本颜色标注
图9:GAN实验训练过程
图11:剖面等距排列
图12:由GAN模型生成的三维空间模型
图13:由ANN模型生成三维空间
图14:由GAN模型生成的三维空间
全文链接:请点击下方的“阅读原文”
·
往期推送
学术论文|基于生成对抗网络的校园总平布局生成式设计探索——以小学校园为例
学术论文|基于神经风格迁移的城市形态生成设计探索——以Tarragones的城市设计为例
喜讯|2021毕设荣获亚洲设计学年奖银奖