Centre for Fine Print Research University of the West of England Centre for Fine Print Research
 

Durer’s Rhinoceros - Artists’ Approaches to Reproducing Texture in Art


Author:
Parraman, C.
Name of conference:50th Annual Convention of the Artificial Intelligence and Simulation of Behaviour (AISB)
Place: University of Goldsmiths, London
Date: 1-4 April 2014
Paper Presented by: Parraman, C
Published by: Goldsmiths http://doc.gold.ac.uk/aisb50/

Abstract:
Since the introduction of computers, there has been a desire to improve the appearance of computer generated objects in virtual spaces and to be able to display the objects within complex scenes exactly as they appear in reality. This is a straightforward process for artists who through the medium of paint or silver halide are able to directly observe from nature and interpret and capture the world in a highly convincing way.

However for computer generated images, the process is more complex, and could be compared to the notion of Durer’s Rhinoceros – Durer created an image of a rhinoceros based on a description; he had not seen a real rhinoceros but managed to create a convincing likeness. In comparison, if data were inputted to build a rhinoceros on a computer, the accuracy of the image would also be dependent on the appropriateness of the data. Furthermore the computer has no capability to compare whether the rendering looks right or wrong – only humans can make the final subjective decision. I have used the term appropriate, because too much data could also be considered as a potential hindrance: too much information could slow down image processing, too little information could result in an incomplete image. As humans we use highly complex terms to describe the things we see, and which are based on our background, age, education and cultural influences.

My other reason for using Durer’s Rhinoceros, is that the drawing was beautifully textured. Whilst it is not an accurate representation, it captures the roughness and bumpiness of the skin, the layers or folds of hide, the furriness of the ears. However, it is a difficult task to accurately convey all the essential textural elements. In order to translate between subjective and objective, to extrapolate numerical data from natural objects, and present ways that most people can understand is a challenge for many fields and industries. Mathematical models and methods have been developed, but there is an element of ambiguity, adjective and comparison. The evolving question is, what are the elements of paintings produced by artists that capture the qualities, texture, grain, reflection, translucency and absorption of a material, that through the application of coloured brush marks, demonstrate a convincing likeness of the material qualities of wood, metal, glass and fabric? This paper aims to look at the relationship between texture, objects and artists’ approaches to reproducing texture in their art.