Visual information has a strong influence on human behavior and it is important to confirm how building and urban form looks like when planning architectural and urban design. Although physical models or perspective images are used to confirm the appearance of space in general, these methods can be said to be sensory and qualitative evaluation methods. Based on the above, the authors aim to develop a method to quantitatively obtain the appearance of surrounding spatial components inside and outside the building, and to apply it to various kinds of spatial analysis.
The tool for analyzing geometric appearance feature that the authors developed in previous research can obtain apparent size and visual distance for each type of visual object. However, despite the presence of discrete multiple visual area units in the visual area to one type, the tool could not subdivide them. In this research, we propose a method for subdivision of visual area and automatically visualizing omnidirectional detailed visual area constitution.
The flow of a case study for verifying the proposed subdivision method in this research is as follows.
Step-1: In order to obtain the features of the geometric appearance, we use geometric calculation with 3D model using gaze vector array by Fibonacci lattice.
Step-2: We propose the subdivision method as a combination of three processing; type division, object ID division, and area division.
Step-3: We evaluate four types of their combination procedure; (1) type-ID division, (2) type-area division, (3) type-ID-area division, and (4) type-area-ID division.
We confirmed that (3) type-ID-area division is the most effective in subdivision of visual area units through the case study. However, since non-negligible errors occurred, we proposed the way to correct errors and confirmed its effectiveness.
From the above results, we could propose a method for computing and visualizing detailed geometric appearance feature using architectural 3D model.
