Edge-Based Shading as a Depth Cue in Paintings

in Art & Perception
Restricted Access
Get Access to Full Text
Rent on DeepDyve

Have an Access Token?



Enter your access token to activate and access content online.

Please login and go to your personal user account to enter your access token.



Help

Have Institutional Access?



Access content through your institution. Any other coaching guidance?



Connect

We explored how an artist who uses a particular monochrome modern painting style generates the impression of relief in paintings. Three portraits, painted after model, were created especially for the experiment. Photographs of the paintings were presented on a computer screen. To investigate the perceived relief of observers we used a gauge figure task. We expected an effect of background contrast on perceived total depth range of the relief, because this is well known in the case of photography. We found that the contrast with the color of the canvas, white, gray or black, influences the perceived articulation of the relief but does not influence the perceived total depth range of the relief. The major difference between photographs and these paintings is that contrasts in the paintings are built up through edge-based shading, whereas photographs mostly contain tonal-area shading. The classical shape from shading cue does not apply to the impressions of depth evoked by the paintings. Perhaps surprisingly edge-based shading can be as effective as classical ways of creating pictorial relief.

Edge-Based Shading as a Depth Cue in Paintings

in Art & Perception

Sections

References

AdamsA. (1952). Natural-Light Photography. Morgan and MorganHastings-on-Hudson, NY, USA.

BloomerC. M. (1989). Principles of Visual Perception. Design PressNY, USA.

DoernerM. (1971). Malmaterial und Seine Verwendung in Bilde. Ferdinand Enke VerlagStuttgart, Germany.

FehrenbachF. (2002). Der oszillierende Blick. Sfumato und die Optik des späten LeonardoZ. Kunstgesch. 65522544.

FoleyJ. D.van DamA. (1983). Fundamentals of Interactive Computer Graphics. Addison-WesleyReading, MA, USA.

GleizesA.MetzingerJ. (1912). Du “Cubisme” Eugène Figuière Éditeurs Collection “Tous les Arts” Paris France.

GombrichE. H. (1950). The Story of Art. PhaidonLondon, UK.

HockneyD. (2006). Secret Knowledge Rediscovering the Lost Techniques of the Old Masters. Penguin GroupNY, USA.

KoenderinkJ. J.van DoornA. J.KappersA. M. L. (1992). Surface perception in picturesPercept. Psychophys. 52487496.

KoenderinkJ. J.van DoornA. J.KappersA. M. L. (1994). On so-called paradoxical monocular stereoscopyPerception 23583594.

KoenderinkJ. J.van DoornA. J.KappersA. M. L.ToddJ. T. (2001). Ambiguity and the ‘mental eye’ in pictorial reliefPerception 30431448.

KoenderinkJ. J.van DoornA. J.ArendL.HechtH. (2002). Ecological optics and the creative eye in: Perception and the Physical World: Psychological and Philosophical Issues in PerceptionHeyerD.MausfeldR. (Eds) pp.  271304. Wiley & SonsChichester, UK.

KoenderinkJ. J.van DoornA. J.WagemansJ. (2013). SFS? Not likely!i-Perception 4299302.

KoenderinkJ. J.van DoornA. J.AlbertazziL.WagemansJ. (2015). Relief articulation techniquesArt Percept. 3151171.

MayerR. (1951). The Artist’s Handbook of Materials & Techniques. Faber and FaberLondon, UK.

PalmerS. (1999). Vision Science: Photons to Phenomenology. Bradford Books, MIT PressCambridge, MA, USA.

PirenneM. H. (1970). Optics Painting and Photography. Cambridge University PressCambridge, UK.

RamachandranV. S. (1988a). Perception of shape from shadingNature 331163166.

RamachandranV. S. (1988b). Perceiving shape from shadingSci. Am. 2597683.

RogersL. R. (1969). Sculpture Series “The Appreciation of the Arts 2”. Oxford University PressLondon, UK.

van HoutN. (2005). Functies van Doodverf met Bijzondere Aandacht voor de Onderschildering en Andere Onderliggende Stadia in het Werk van P. P. Rubens Thesis Katholieke Universiteit Leuven Belgium.

von HildebrandA. (1913). Das Problem der Form in der Bildenden Kunst. HeitzStrassburg, Germany.

Figures

  • View in gallery

    Marianne Venderbosch, Model on black canvas, 2008, acrylic on canvas, 120 × 100 cm. Notice that the ground color is part of the image. Photograph by Marianne E. Venderbosch. © Marianne E. Venderbosch.

  • View in gallery

    The object, a sphere, is identically shaded in all three images. In the middle image the ground color is much lighter than the object, in the image at right the ground color is much darker. In these cases the contrast polarity between the object and the ground color is fixed. In the left image the ground color is neither darker nor lighter than the object. At the top the object is lighter than the ground color, at the bottom it is darker. At middle and right the contrast switches polarity. Notice that the object in the left image has the highest perceived relief.

  • View in gallery

    (Top) We compare a simple form of edge-based shading as articulated from gray, white, and black grounds. The articulation is similar in all three cases; apparently the tone of the ground is almost irrelevant. Notice that the ‘ground’ is not ‘background’ in the sense of ‘backdrop’. Although there certainly is a sense of figure–ground segregation, especially in the center figure, the ground does not strongly appear to continue ‘behind’ the figure. In the middle and right figures it perhaps does so more. In all three cases there is no sense of a background of any specific shape — say surface like — or any specific depth. (Bottom) We compare the influence of shading contrast (left) and edge blur (right) on the articulation induced by the conventional shading stimulus (center; Ramachandran, 1988a, b). Apparently both shading contrast and edge blur have a strong flattening effect on the articulation. The blurred figure seems to grow out of the ground color without much of an apparent depth gap. The low contrast figure is definitely less ‘in front of’ the ground color than the high contrast one. In neither case is the ground color clearly defined as a ‘surface’, nor at any specific ‘depth’ with respect to the figure.

  • View in gallery

    Paintings created by Marianne Venderbosch as stimuli for the experiment. Acrylic on canvas. In the left painting the face is modulated from the gray surface with white and black paint, in the middle the face is modulated from the white surface with nearly only black paint and at right the face is modulated from the black surface with mostly white and a little black paint. Photographs by Marianne E. Venderbosch. © Marianne E. Venderbosch.

  • View in gallery

    At left a schematic representation of pictorial space. The picture plane is not represented, since it is not in pictorial space. The ‘background’ is symbolically represented, but it is not necessarily surface-like, nor at some well defined depth with respect to the relief of the face. Everything is related to the direction of view. The ‘relief’ represents the ‘pictorial face’. The ‘pictorial head’ has only a face (front), but not a backside. The background is often experienced as ‘continuing behind the face’, although the depth gap is fully ambiguous. The relief can be described in terms of a global egg-shape (technically a quadric), and a superimposed articulation. In the inset at right we show the articulation as the difference between the relief and a quadric approximation (dashed curve). The ‘depth range’ is defined as the total depth range subtended by the relief (A in the figure). A measure of the strength of the articulation is the amplitude of the deviation from the global egg shape (B in the figure). In the experiment we can only measure the relief, the ‘depth’ of the background is not accessible.

  • View in gallery

    This figure illustrates the use of the gauge figure. We have superimposed two gauge figures over the face in a portrait. The one in white can be seen to ‘fit’ the relief of the cheek whereas the one in black does not fit at all and is not perceived as being ‘on’ the face. Notice that the ‘fit’ is not a physical fit but depends on the awareness of the observer. The axle, that is seen to stick out from the plane of the circle, is used to indicate the frontal side of the disk, and thus to disambiguate the foreshortening of the circle. Photograph by Marianne E. Venderbosch. © Marianne E. Venderbosch.

  • View in gallery

    (Left) The perceived total depth ranges of the relief as a function of the observer, sorted with respect to the mean. The drawn line denotes the gray, the dashed line the white, and the dotted line the black tonal ground. (Right) The perceived total depth ranges of the relief as a function of the ground tonal value. For the gray stimulus the sequence from top to bottom is AD, SP, MW, JK, MV, HR. The numbers on the vertical scale represent the total depth range of the relief in arbitrary units (AU).

  • View in gallery

    The articulation of the reliefs for all observers and for all ground tonal values. The dark-to-light scale signifies far-to-close in depth.

  • View in gallery

    (Left) The articulation as a function of the observer, sorted with respect to the mean. The drawn line denotes the gray; the dashed line the white, and the dotted line the black ground tone. (Right) The articulation as a function of the ground tonal value. For the black stimulus the sequence from top to bottom is MV, JK, MW, AD, HR, SP. The numbers on the vertical scale represent the articulation in arbitrary units (AU).

Index Card

Content Metrics

Content Metrics

All Time Past Year Past 30 Days
Abstract Views 10 10 6
Full Text Views 5 5 5
PDF Downloads 0 0 0
EPUB Downloads 0 0 0