I have rarely touched on the nature of CGI and yet it is through computer generated imagery that the majority of people first encounter drawings, especially as so much of children's film production uses computer supported animation. Software to run 3D drawing programs such as Maya is becoming more and more available to artists and therefore, especially as I'm trying to look at certain aspects of perception in more detail, I thought it worthwhile adding my thoughts on what it means to experience images produced by this ever growing area of practice.
T figure: pen and ink on paper
I have no technical experience in this area of work, so the first thing I had to do when I wanted to explore the area, was to find someone who did, and Matt Clark of the Leeds Arts University animation department was the person I went to to give me an overview of what could be done. This was about five years ago and I'm sure he is now using totally different software but I'm presuming the process would be similar if I was to ask him again. I wanted to see if it was possible to create a 3D animated version of my own body and I was told that the first thing to do was to generate a T figure by hand and the drawing above was what I did at the time, pot belly included, as I attempted to be as 'realistic' as possible. Matt then took the drawings away (I think I did a back view as well but couldn't find that) and from it generated these wonderful flattened and opened out 3D grids for me, as well as the figures below it, one version without and a second one, at my insistence with genitals.
Initial figure without genitals
Redrawn with genitals
I found the process interesting and the resultant figure was very close in terms of its proportional visual relationship to my initial drawings but I was very distant in terms of any form of control over what was going on. I knew that the figure would be constructed using some sort of software like Blender, but what I was not prepared for was the degree of 'smoothing' that was going to be used to create a surface for the figure. I felt this was totally at odds with any form of intuitive feeling or emotional engagement I might have. In effect the surface seemed to distance me from the figure, even though I knew that it was based on myself and my measurements.
From a 'Blender' tutorial
As you can see from the 'Blender' tutorial above, a face has been constructed from a front and side view, just as in my initial experience, and the grid that has been pulled over the face to make a head has also been 'filled in' with a plastic effect, shiny surface. In effect for myself, this surface makes it feel as if the figure is made of plastic.
The next two images are from another 'Blender' tutorial, this time of two photographs, and it is part of the tutorial that shows someone how to use 'pins' to locate significant features, such as nose width, eyebrows, chin, forehead and mouth on a frontal view and link these to the same features on a side view. You can see the 'ghost' image of the pre-existing grid on the lower image as it is pulled over the photograph of the head, interesting to see how much movement there is between the grid and the actual face, the 'ghost grid' ear for instance will need to move left to coincide with the photographic ear, whilst the back of the head grid will need to be pulled to the right to coincide with the photograph.
I have only recently begun to understand some of this and as always as I begin to get my head around things I tend to go off on tangents, so bare with me as I ramble around, because in order for me to make sense of something to myself, I have to get a feeling for the various entanglements that something has with the rest of this world. In this way I can see it more as a process rather than as a thing. I can also attempt to resolve the relationship between what I am feeling about something and what I am thinking about.
A recent project on perception finds me working alongside other members of a project team, including some who are working using computer generated imagery. Safa Tharib in particular has highlighted for me some aspects relating to the history of drawing in this area, and I'm hoping he will forgive me my re-rendering his Powerpoint in with my own words below. I'm trying to bridge that gap between a computer science understanding and my own very fine art, hand made understanding of these issues, i. e. attempting to answer my own self set questions as to how and why I might use computer aided drawing to make a series of images related to how the perception of our inner bodies can lead to the creation of images that can be both expressive of my own feelings about these things and be capable of communicating other people's thoughts and feelings about the same perceptions.
Before computers technical designs were drawn using rulers, compasses and protractors, but freeform curves such as ones used to visualise a ship's bow, had to be drawn using other means. In my day when I worked in a drawing office, we used what we called a 'Flexicurve' and in earlier times they used to use flexible strips of wood, called 'splines'. The Oxford English Dictionary finds the first recorded usage of the word 'spline' in the 18th century in East Anglia, England, and suggests the term spline may be related to splinter, something that can easily happen when you bend a thin strip of wood too far. As well as to help shape the curve of a ship's prow, spline devices have been used to design shapes for pianos, violins, and other wooden instruments, and the Wright brothers used one to shape the wings of their aircraft.
A flexicurve
A spline being made
Using 'ducks' to control a thin wooden strip's curvature
I have reflected on the use of aesthetically satisfying curves in earlier posts, so perhaps at this point it might be useful to go back and look at these issues, because they highlight the interrelationship between the physical, the mathematical and the conceptual, in particular when we look at the physical way that these splines are produced and how curves are anchored in vector drawing software packages. In AutoCAD for instance, a spline is defined as " a smooth curve that passes through or near a set of fit points, or that is defined by the vertices in a control frame." This reminds me of earlier posts on euler spirals and the Bézier Curve. Literally tangents that can be followed, but which are also concepts that entangle themselves with a wider set of concepts, including the spiralling of the winding thread as in clothoid spirals and the way human beings need to be subtlety visually controlled as they drive cars.
How to work with splines in AutoCad
There is something really satisfying about curves that are defined by constraints, and perhaps it is because we sense those underlying physical constraints that we feel a certain visual satisfaction when we encounter them.
Ducks
Splines were held in place at a number of predetermined points by weighted objects called ‘ducks’ and between the ducks the elasticity of the spine material caused the strip to take the curvature that minimised the energy of bending, thus creating the smoothest possible constrained shape of curvature, a curve that could furthermore be tweaked by moving the ducks.
Architects are of course also very interested in these issues, and it is in architectural theory that another series of associations is opened out, in this case an association with the psychoanalytic theories of Freud, Andrew Benjaminan providing an unexpected link to a thread that gives myself an 'excuse' to draw images that reflect on that close relationship between inner feelings and external 'objective' reality.
Andrew Benjamin lecture
In Freud's case he is looking for a language to describe a loss or change of energy, so he uses the metaphor of a wave and a wave becoming stilled. A change in modulation or frequency of an energy as it moves from one area to another is seen as what drives certain mental issues, such as melancholia. (Not that Freud really gets to grips with this idea in his diagram, but the idea of energy transfer is already helping me to think of my own recent drawings as places where energy transfer can take place between diagrams and representations) Benjamin highlights the idea of the divided self, a self that is part conscious and part unconscious, between which there is also constant movement, some things moving from conscious experience into the unconscious and some things moving from the unconscious into consciousness. We have no power over what is emerging from the unconscious, but we are able to control what emerges from conscious thought and it is in this divided but connected situation that Benjamin further argues that we need to be thinking about how a drawn line both presents a possibility and exists as a physical entity in its own right. We are once again looking at process here and indeed it could therefore be argued that what we think of as the self is really the after effect of a process. The drawing by Freud below being a static image of a dynamic process. It's lines create an order that moves around an almost Cartesian x and y axis; one thing being linked to another and this sequential system also means that we can read this drawing as a type of narrative, a type of narrative that is generative, i. e. that we can create further images that are reflections on the issues that Freud is mapping out in this and other drawings.
Freud's drawings
Looking at the drawings above, we can easily see the differences between an early freehand diagram and later technical drawings, used by Freud to illustrate his ideas. It is fascinating to see how a difference in construction effects our understanding. The freehand drawing suggests a mind roughing out an idea, it is a drawing that presents the development of a concept that is yet to be fully decided upon. Whilst the precise geometry of the later diagrams, produces a set of drawings that present a much firmer commitment and clarity of idea.
So what ties these things together? They are all concerned with ideas about our existence in a world that consists of both an external 'surface' reality and an inner underlying, often unconscious world of feeling and intuition. There are, I would argue, some basic elements that are needed to construct images, whether in a 3D computer programme these be NURBS or polygons, or in a human brain, where there resides image schemas, those small units of information that inform basic, automatic actions, and that create neuronal patterns, which are the abstractions that form the basis of our ability to undertake real-time actions. In relation to which, the cognitive psychologist Raymond Gibbs wrote: "Image schemas represent the essential glue that binds embodied experience, thought, and language", and I would argue that the elements that make up a net, (whether constructed as NURBS or polygons); those points and lines and planes, are the basic 'bits' that I use when I make a drawing and which I can only use because I have access to image schemas of my own, based upon the fact that I inhabit a particular type of body.
Going back to the spline, we have an idea of points of control, (consciousness) and the idea of effect (the various curves). If we look at the 'Blender' tutorial using pins to link front and side views, we can also in our minds hold two images, on the one hand it could be a NURBS defined net, of a spline being made and the ducks being points of control or if it is a polygon modelling programme, of squares and triangles being joined together at different angles. Both these images are revealed as a net, which is designed as if it is a sort of memory form of an average three dimensional face. This face is made more and more 'like' the 2D images over which it is plotted; the more pins or points put in place, the more the net can be pulled into a particular shape, and the less points are pinned the more the mesh of points, lines and shapes finds its own form. Going back to what Benjamin was talking about, the drawn line both presents a possibility and exists as a physical entity in its own right.
So what has this got to do with perception? I would argue that it is the idea of 'potentiality' or 'possibility' that binds together the moment of perception and an attempt to represent or image it. As argued previously, the body needs to react almost instantaneously to changing stimulus, both from inside and outside itself, if it is to react in a timely manner to external change. The construction of an animated figure and the establishing of a convincing external 'skin' or surface, is dependent on the construction of forms using a geometry that can control constantly reforming nets that both describe 3D surfaces and which can eventually become them. (As when 3D printed for example). However in order that the system doesn't get too complicated we tend to use attractors, or preferred patterns to create stability within complexity. NURBS which as I have recently found out are used to create interactive surfaces for 3D modelling, would appear to work in this way. Robin Evans' idea that meaning is the secondary effect of the operation of geometry, in this case in relation to computer generated imagery, suggests to myself that it is the net itself that creates meaning in relation to the forms it makes.
So how does it do this? First of all it would be useful to read an earlier post on cross contour drawing. In this post I suggest that this form of drawing can be a very powerful way of visualising mass and that it is also a way that touch can be visualised, i. e. that it is a synaesthetic mode of drawing capable of combining sight and touch in its realisation. This is why sculptors in particular like to use it, as it can carry information about the solidity of the experience of making things. A few years ago I made some cardboard models of fires. I took photographs from various positions and loaded them into 123DCatch, a 3D image making program that was at the time free and easy to use. (It has since been discontinued) I was interested in the image making potential of the software and wanted to explore the idea of making a solid something that was in constant movement. In effect to create a 'mountain' out of a gas, especially as I was working in one medium that had no substance beyond a mathematical code of zeros and ones and another that was seen as cheap packaging material and usually put out for recycling.
Virtual fires
Some images I left as they emerged and others I began to process, but in all cases I saw the distorted 3D grid as the main carrier of the message. The grid was in effect what I believed gave the images 'conviction'. Then a little later because of my growing interest in CGI I began my collaboration with Matt Clark and these moving images, based on drawings of myself, were produced.
Matt Clark: Joint retargeting test
Matt Clark (Chicken man)
I found it hard to associate with the images created, and unlike the fire images whereby I could see how the 3D grids gave a certain gravity to the forms, felt nothing beyond a certain humour in the way my surrogate body could be made to move. Matt making me laugh at what could be achieved, but like most jokes on extended play they tend to lose their potency.
Going back to the group research on perception, hopefully this rambling post can operate as an opening gambit, that begins to explain how I might be able to link with, interject or comment on the work of the other people in the group. It also hopefully serves to illustrate another issue, that of self reflection during the process of developing new bodies of work, something that those of you who are students, are constantly being asked to do.
Work in progress; stages in image development of perception of sexual arousal.
(The conjunction of diagram and representation)
References:
Robin Evans (2000) The Projective Cast: Architecture and Its Three Geometrics MIT Press
Gibbs, R.W., (2005) The psychological status of image schemas. In From perception to meaning: Image schemas in cognitive linguistics, 29, pp.113-136. Mouton de Gruyter press
See also:
Qualia (Part one of a series of posts related to perception, of which this post is part two)
FAT is dead
Virtual reality
3D Printing
Data visualisation
Artificial intelligence
Virtual space and metaphor
Virtual reality
3D Printing
Data visualisation
Artificial intelligence
Virtual space and metaphor
No comments:
Post a Comment