【作者】 臧伟； 郭亚楠； 周鸣浩； 王信； 李伟伟； 朱赟；

【Author】 Zang Wei;Guo Yanan;Zhou Minghao;Wang Xin;Li Weiwei;Zhu Yun;Key Laboratory of Ecology and Energy Saving Study of Dense Habitat,Ministry of Education,Tongji University;College of Architecture and Urban Planning Tongji University;Joint International Research Laboratory of Eco-Urban Design (Tongji University),MINISTRY OF EDUCATION;Tongji Architectural Design (Group) Co,.Ltd.;

【机构】 同济大学高密度人居环境与节能教育部重点实验室； 同济大学建筑与城市规划学院； 教育部生态化城市设计国际合作联合实验室(同济大学)； 同济大学建筑设计研究院(集团)有限公司；

【摘要】 在城市设计中需要考虑人主观因素的比重较大,但是主观因素又很难转换为定量的数据进行分析。我们设计出一种基于VR眼动数据采集和分析方法,为设计师获取可靠的主观数据提供一套可行性强且廉价的解决方案。以喀什高台民居保护与设计项目为例,方法的实现和应用流程为:首先搭建虚拟现实场景,通过倾斜摄影、全景拍摄、建模等多种方式结合,搭建了高台民居现状的三维场景;然后根据设计团队的CAD模型搭建设计方案的三维场景;最后在通过Unity3D引擎进行眼动仪的软硬件对接、VR场景交互设计、程序设计、相关算法实现。实际应用中通过虚拟现实头盔和眼动仪采集被试人员的眼动数据,并设计数据的分析方法:从眼动数据中分别筛选出被试人员的注视时间、注视范围、注视频次,通过算法使数据可视化。通过该方法,实现高台民居的现状和设计方案的实时比对;通过眼动数据分析,帮助设计团队更方便的筛选出高台民居需要保留的建筑风貌,为高台民居保护设计方案的优化提供数据支撑;整个方法具有较强的可推广性。更多还原

【Abstract】 Subjective factors need to be considered in urban design,while they are difficult to quantify and analyze.In this paper,the authors propose a method based on VR eye-movement data collection and analysis to provide a set of feasible and inexpensive solutions for the designers to obtain reliable data.Take the Preservation and Design Project of Hathpace Houses in Kashi as an example,this paper introduces the implementation process of this method as followed.First of all,build a VR scene:Build a 3 D current situation scene of Hathpace Houses by integrating tilt-photography,panoramic shooting,modeling and other methods.Then,3 D scene of the design proposal is built based on the CAD model provided by the design team.Finally,Unity3 D engine is used to connect the software and hardware of eye tracker,and for interaction design of VR scenes,programming,algorithm implementation.In practices,eye-movement data of the experimenter is collected by Helmet-Mounted Displays(HMD) and eye tracker.Data analysis method is designed as followed:filter the eye-movement data to get experimenter’s gaze time,gaze range and gaze frequency;use entropy weight method to integrate these three types of data and visualize data through algorithms.In this way,the real-time comparison between current status and proposed design scenario of Hathpace Houses can be achieved.The eye-movement data make it easier for the design team to select the architectural features of Hathpace Houses that need to be preserved,which could provide data to support the preservation proposals.The whole method has strong generalizability.更多还原