| Summary: | 碩士 === 國立臺灣科技大學 === 建築系 === 91 === Clay molding is an easy way to form 3D objects. It is often used in early architectural design phases to conduct massing and form studies because it allows designers to create and modify 3D objects freely and rapidly. As Schön (1983) explains in reflective practices, the created objects, in turn, stimulate designers to further their concept development. This thesis is inspired by the Electronic Cocktail Napkin, which uses sketch as the driving force to associate other visual references (Gross, 1996). The aim of the thesis is to create a computing environment that uses clay as the interface to access libraries of visual references to aid massing and 3D form studies in early design phases.
Clay Form Case Finder is a proposed environment to assist designers by promoting reflection in actions. Clay Form Case Finder consists of a working surface, a display monitor, video cameras, and a database. Designers mold shapes out of clay that produce 3D objects on the working surface. Inputs from video cameras are processed to extract specific features: shape primitives, element count, and spatial relations. These features are then used to match against features of design precedents in the database. Precedents selected from the database are shown on the display monitor.
The thesis implements a Clay Form Case Finder prototype. Due to the time constraint, the feature extraction process is performed manually (i.e., by a person). In addition, an experiment is conducted to examine how such environment may assist expert and novice designers. Results from the experiment suggest that the Clay Form Case Finder prototype stimulates design reflections as intended. Furthermore, the feature matching mechanism, although naive, produces search results that agree with designer’s expectations.
Clay Form Case Finder is a feasible environment to promote 3D form studies through visual case feedbacks. The thesis demonstrates this result through an implemented prototype and user experimentations. To achieve the ideal Clay Form Case Finder, however , two continuing efforts are needed: first, to automate the feature extraction process; second, to enrich the database of visual references.
|
|---|
