Representing Style by Feature Space Archetypes: Description and Emulation of Spatial Styles in an Architectural Context

Style is a broad term that could potentially refer to any features of a work, as well as a fluid concept that is subject to change and disagreement. A similarly flexible method of representing style is proposed based on the idea of an archetype, to which real designs can be compared, and tested with examples of architectural plans. Unlike a fixed, symbolic representation, both the measurements of features that define a style and the selection of those features themselves can be performed by the machine, making it able to generalise a definition automatically from a set of examples. At its core, style is what distinguishes one group of works from another. This paper proposes that we can define a style using an archetype, an ideal model comprised of the features that exemplify the style. This concept differs from the description of a type, or category into which particular examples can fall, and from that of a prototype, precedent or case, which are actual instances on which later examples can be modelled. An archetype is something between the two, a generalisation that can not exist materially, yet matches and is compared to many actual instances. This is almost certainly not a real example, but an abstraction made up of only those features necessary to differentiate it from other archetypes. Many approaches to style are based on explicit symbolic representations (where fixed concepts are mapped to named variables) or rule systems. These can tell us useful things about designs and how they can be made, but are inflexible. They reveal some of the ways we learn about styles pedagogically, but typically fixed, historical ones. By contrast this work proposes a method to automatically derive representations from real examples of design. It is based on the mapping of design examples in a high dimensional feature space, and uses methods of dimensionality reduction of this space to yield an archetype that describes the style. This can be used to classify, and

[1]  John S. Gero,et al.  A qualitative feature-based characterization of 2D architectural style , 2006, J. Assoc. Inf. Sci. Technol..

[2]  George Stiny Two Exercises in Formal Composition , 1976 .

[3]  Kimberle Koile,et al.  Design Conversations with Your Computer: Evaluating Experiential Qualities of Physical Form , 1997 .

[4]  David G. Stork,et al.  Pattern Classification , 1973 .

[5]  Alan Penn,et al.  Natural Movement: Or, Configuration and Attraction in Urban Pedestrian Movement , 1993 .

[6]  William J. Mitchell,et al.  The Palladian Grammar , 1978 .

[7]  Hugh Durrant-Whyte,et al.  Autonomous Navigation in Unstructured Environments , 2004 .

[8]  A. Turner,et al.  From Isovists to Visibility Graphs: A Methodology for the Analysis of Architectural Space , 2001 .

[9]  Shigeo Abe DrEng Pattern Classification , 2001, Springer London.

[10]  Antonio Robles-Kelly,et al.  Edit distance from graph spectra , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[11]  Nelson Goodman,et al.  The Status of Style , 1975, Critical Inquiry.

[12]  Alasdair Turner,et al.  An Algorithmic Definition of the Axial Map , 2005 .

[13]  Murray Silverstein,et al.  A Pattern Language , 1977 .

[14]  M. Turk,et al.  Eigenfaces for Recognition , 1991, Journal of Cognitive Neuroscience.

[15]  Richard Freedman,et al.  Possible Palladian Villas: Plus a Few Instructively Impossible Ones , 1992 .

[16]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[17]  K. Pease,et al.  Secure foundations : key issues in crime prevention, crime reduction and community safety , 2000 .

[18]  L. B. Alberti,et al.  Ten books on architecture , 1956 .

[19]  John S. Gero,et al.  ENTROPIC-BASED SIMILARITY AND COMPLEXITY MEASURES OF 2D ARCHTECTURAL DRAWINGS , 2001 .

[20]  John S. Gero,et al.  SKETCHES AS AFFORDANCES OF MEANINGS IN THE DESIGN PROCESS , 2001 .

[21]  Barbara Tillmann,et al.  MUSICAL STYLE PERCEPTION BY A LINEAR AUTO-ASSOCIATOR MODEL AND HUMAN LISTENERS , 2004 .

[22]  F ROSENBLATT,et al.  The perceptron: a probabilistic model for information storage and organization in the brain. , 1958, Psychological review.

[23]  E. Gombrich Art and Illusion , 1962 .

[24]  Vladimir Cherkassky,et al.  The Nature Of Statistical Learning Theory , 1997, IEEE Trans. Neural Networks.

[25]  H. Koning,et al.  The Language of the Prairie: Frank Lloyd Wright's Prairie Houses , 1981 .

[26]  Teuvo Kohonen,et al.  Self-organized formation of topologically correct feature maps , 2004, Biological Cybernetics.

[27]  B. Hillier,et al.  The Social Logic of Space , 1984 .

[28]  Ruth Dalton,et al.  Small-Graph Matching and Building Genotypes , 2008 .

[29]  Kimberle Koile,et al.  An Intelligent Assistant for Conceptual Design , 2004 .

[30]  Wrg Hillier,et al.  Crime and urban layout: the need for evidence , 2000 .

[31]  G. Stiny Introduction to Shape and Shape Grammars , 1980 .

[32]  J. Gibson The Ecological Approach to Visual Perception , 1979 .

[33]  Richard Weston Key buildings of the twentieth century : plans, sections, and elevations , 2004 .