|
Home | Project Description | People | Publications | Links | Events | Contact |
||
|
Research questions | Research context | Research methods | References | Examples of shape computations |
||
|
|
||
|
|
Research Context Shape and form are perhaps the most influential characteristics in the perception of design outputs such as sculpture, clothing, and consumer products. While various forms of 2D and 3D representations might be exploited in the shape generation process, sketching still forms a key technique for practitioners. Design sketches of shape and form are a particularly interesting category of visual representation because perception and interpretation are bound together with creation and evaluation in the shape sketching process. Fish [1996] proposed that indeterminacy or vagueness in sketches exploits a capability of the human brain to make sense of incomplete information. He linked the physical markmaking act to cognitive mechanisms in his proposition that ambiguity may have a positive and deliberate function in design sketches. Sketching appears to have a direct relationship with seeing and ‘selecting’ in the emergence of design ideas through interpretative and transformational processes [Soufi et al 1996,Goel 1995]. Studies of cognitive processes in design [Akin 2001] and shape cognition [Wang 1998] have identified differences between design disciplines. Wang’s studies, for example, found that shape perception tendencies between architects, industrial designers and graphic designers are different. Architects reveal an ability for identifying emergent shapes associated with transformational processes, whereas industrial designers had a superior ability for interpreting volumetric sub-shapes from two dimensional representations. Thus studies of sketching are viewed as potentially revealing for this project. Over the last thirty years, efforts have been made in various disciplines, including art and design, architecture, and product design, to apply shape grammars to the analysis of styles and the generation of design families. Chau [Chau 2004] provides a comprehensive diagram (reproduced in Figure 3) plotting existing applications along a timeline. These have demonstrated the viability of generative techniques to capture and reproduce styles in a range of design domains. |
|
|
|
Figure 3. Shape grammar applications in design
|
|
|
|
This project will explore the use of shape grammars to realize what Smyth and Wallace [Smyth et al 2000] refer to as a “form synthesis engine” within their model for the synthesis of aesthetic product form. Some progress has been made towards this goal. Tapia [1999] demonstrated a robust implementation using straight lines in 2D and Chau et al [2004] describe a 3D shape grammar implementation using both rectilinear and curvilinear elements. Jowers et al. [Jowers et al 2004] report success in 2D Bezier curves. A key problem in all of these systems lies in the detection of the sub-shapes to be used in the computation of new shapes. For example, in Chau’s system, human intervention is needed to identify the sub-shapes from which further shapes are to be generated whilst Tapia’s system solved the problem for 2D lines but in a way that cannot be extended to the 3D freeform curves that typify today’s consumer product designs. In parallel, researchers in the computer vision community have established a range of techniques that enable the identification of shapes in real-world situations: for example, the use of statistical learning algorithms for modeling and recognizing new object categories [Heap et al 1998, Baumberg et al 1994]. This project will explore the application of the techniques used for the recognition of visual objects to sub-shape detection in shape grammar-based design systems. Coupling learning about how designers generate shapes with understanding of principles of shape computation will provide insights into potential interplays between shape computation technologies and design practices. Even without the experimental prototype, the project will enhance design creativity by enabling the construction of solution spaces - so enabling designers to explore them more extensively and systematically than is presently possible. If we are successful in applying the software prototype to a selection of design cases we will be able to demonstrate how shape computation systems can expand the design space that designers consider, so augmenting design activity and enhancing their creativity. The designers who participate in the project will gain insights on how they generate shapes and this, through reflective practice, is likely to contribute to their personal professional development. For people who are not participants in the project, the outcomes of the research are, in the short term, likely to be of most interest to design educators since understanding of how designers generate shapes could be used to strengthen theoretical underpinnings of design pedagogy. If the investigations into the use of vision recognition techniques provide positive results then CAD system vendors are likely to be interested in the research results. In the longer term, when shape synthesis systems are available on the market, design practitioners will have access to new tools which, by highlighting possible avenues for shape generation, will enhance their creativity.
|
|
|
Research questions | Research context | Research methods | References | Examples of shape computations |
||
|
|
|
|
|
Department of Architecture |
Faculty of |
Department
of Design |
