A model for the rapid interpretation of line drawings in early vision

Abstract According to the prevailing view, the elements of early visual processing arecharacterized by simple geometric properties such as length, orientation, andcurvature. We demonstrate in this chapter that this view must be revised—theelements of early vision need not be geometrically simple. Instead, they can becharacterized in terms of environmental relevance, computational architecture,and processing speed. We begin by summarizing the conventional view of earlyvision and point to several problems it encounters. We then re-examine therole played by the elements of early vision, arguing that it is advantageous forthem to describe environmentally relevant properties, even if these quantitiesare not always valid. As an illustration, we develop a computational model forthe rapid recovery of one important scene property from line drawings—thethree-dimensional orientation of objects. Data from recent visual searchexperiments in humans are presented in support of the model. Early vision

[1]  David L. Waltz,et al.  Generating Semantic Descriptions From Drawings of Scenes With Shadows , 1972 .

[2]  Anne Treisman,et al.  Features and objects in visual processing , 1986 .

[3]  James T. Enns,et al.  Three-dimensional features that pop out in visual search. , 1990 .

[4]  I E Holliday,et al.  SEARCH FOR STEREOSCOPIC SLANT DIRECTION IS PARALLEL , 1989 .

[5]  Jan A. Mulder,et al.  Reconstructing polyhedral scenes from single two-dimensional images: the orthogonality hypothesis , 1990 .

[6]  W. Epstein,et al.  In search of depth , 1990, Perception & psychophysics.

[7]  G W Humphreys,et al.  Grouping processes in visual search: effects with single- and combined-feature targets. , 1989, Journal of experimental psychology. General.

[8]  D L Butler,et al.  Integration of features in depictions as a function of size , 1987, Perception & psychophysics.

[9]  H. Nothdurft Sensitivity for structure gradient in texture discrimination tasks , 1985, Vision Research.

[10]  Alan K. Mackworth Interpreting Pictures of Polyhedral Scenes , 1973, IJCAI.

[11]  Jon Driver,et al.  Visual search for a conjunction of movement and form is parallel , 1988, Nature.

[12]  A Treisman,et al.  Feature analysis in early vision: evidence from search asymmetries. , 1988, Psychological review.

[13]  S. Sutherland Seeing things , 1989, Nature.

[14]  B. Julesz A brief outline of the texton theory of human vision , 1984, Trends in Neurosciences.

[15]  Ronald A. Rensink,et al.  Sensitivity To Three-Dimensional Orientation in Visual Search , 1990 .

[16]  A. Treisman,et al.  Search asymmetry: a diagnostic for preattentive processing of separable features. , 1985, Journal of experimental psychology. General.

[17]  W. R. Garner,et al.  Visual texture segregation based on orientation and hue , 1986, Perception & psychophysics.

[18]  Alan K. Mackworth,et al.  Model-Driven Interpretation in Intelligent Vision Systems , 1976, Perception.

[19]  T. Callaghan Interference and dominance in texture segregation: Hue, geometric form, and line orientation , 1989, Perception & psychophysics.

[20]  J. Koenderink,et al.  Sine-wave gratings:Scale invariance and spatial integration at suprathreshold contrast , 1983, Vision Research.

[21]  A. Treisman Features and Objects: The Fourteenth Bartlett Memorial Lecture , 1988, The Quarterly journal of experimental psychology. A, Human experimental psychology.

[22]  Y. J. Tejwani,et al.  Robot vision , 1989, IEEE International Symposium on Circuits and Systems,.

[23]  J. Enns,et al.  Seeing textons in context , 1986, Perception & psychophysics.

[24]  A. Macworth Interpreting pictures of polyhedral scenes , 1973 .

[25]  James T. Enns,et al.  Components of Line-Drawing Interpretation: A Developmental Study. , 1990 .

[26]  J. Duncan,et al.  Visual search and stimulus similarity. , 1989, Psychological review.

[27]  S. Ullman Visual routines , 1984, Cognition.

[28]  David S. Johnson,et al.  Computers and Intractability: A Guide to the Theory of NP-Completeness , 1978 .

[29]  Deborah Walters,et al.  Selection of image primitives for general-purpose visual processing , 1987, Comput. Vis. Graph. Image Process..

[30]  Irving Biederman,et al.  Human image understanding: Recent research and a theory , 1985, Comput. Vis. Graph. Image Process..

[31]  James T. Enns,et al.  2 The Nature of Selectivity in Early Human Vision , 1992 .

[32]  Susan L. Franzel,et al.  Guided search: an alternative to the feature integration model for visual search. , 1989, Journal of experimental psychology. Human perception and performance.

[33]  Ken Nakayama,et al.  Serial and parallel processing of visual feature conjunctions , 1986, Nature.

[34]  James T. Enns,et al.  A developmental study of shape integration over space and time. , 1986 .

[35]  Alex Pentland,et al.  Perceptual Organization and the Representation of Natural Form , 1986, Artif. Intell..

[36]  S. Ullman,et al.  Curve tracing: A possible basic operation in the perception of spatial relations , 1986, Memory & cognition.

[37]  D. Perkins Visual discrimination between rectangular and nonrectangular parallelopipeds , 1972 .

[38]  Patrick Cavanagh,et al.  11 – FORM PERCEPTION AND ATTENTION: Striate Cortex and Beyond , 1990 .

[39]  V S Ramachandran,et al.  Perceiving shape from shading. , 1988, Scientific American.

[40]  Ronald A. Rensink,et al.  Influence of scene-based properties on visual search. , 1990, Science.

[41]  D. Badcock,et al.  Processing feature density in preattentive perception , 1988, Perception & psychophysics.

[42]  Michael Kubovy,et al.  The Psychology of Perspective and Renaissance Art. , 1986 .