Object Perception: Generative Image Models and Bayesian Inference

Humans perceive object properties such as shape and material quickly and reliably despite the complexity and objective ambiguities of natural images. The visual system does this by integrating prior object knowledge with critical image features appropriate for each of a discrete number of tasks. Bayesian decision theory provides a prescription for the optimal utilization of knowledge for a task that can guide the possibly sub-optimal models of human vision. However, formulating optimal theories for realistic vision problems is a non-trivial problem, and we can gain insight into visual inference by first characterizing the causal structure of image features-the generative model. I describe some experimental results that apply generative models and Bayesian decision theory to investigate human object perception.

[1]  N. Kanwisher,et al.  The lateral occipital complex and its role in object recognition , 2001, Vision Research.

[2]  M. Landy,et al.  Measurement and modeling of depth cue combination: in defense of weak fusion , 1995, Vision Research.

[3]  R. Jacobs.  What determines visual cue reliability? , 2002, Trends in Cognitive Sciences.

[4]  Daniel Kersten,et al.  High-level Vision as Statistical Inference , 1999 .

[5]  H H Bülthoff,et al.  Integration of depth modules: stereo and shading. , 1988, Journal of the Optical Society of America. A, Optics and image science.

[6]  Tomaso Poggio,et al.  Computational vision and regularization theory , 1985, Nature.

[7]  Eero P. Simoncelli.  Statistical models for images: compression, restoration and synthesis , 1997, Conference Record of the Thirty-First Asilomar Conference on Signals, Systems and Computers (Cat. No.97CB36136).

[8]  Rajesh P. N. Rao,et al.  Probabilistic Models of the Brain: Perception and Neural Function , 2002 .

[9]  Paul R. Schrater,et al.  Pattern inference theory: A probabilistic approach to vision , 2002 .

[10]  M. Gazzaniga.  The new cognitive neurosciences, 2nd ed. , 2000 .

[11]  U. Grenander.  Elements of Pattern Theory , 1996 .

[12]  Maggie Shiffrar,et al.  The influence of terminators on motion integration across space , 1992, Vision Research.

[13]  Brian D. Ripley,et al.  Pattern Recognition and Neural Networks , 1996 .

[14]  T. Hendler,et al.  Object-completion effects in the human lateral occipital complex. , 2002, Cerebral cortex.

[15]  D. M. Green,et al.  Signal detection theory and psychophysics , 1966 .

[16]  D. Mumford.  On the computational architecture of the neocortex , 2004, Biological Cybernetics.

[17]  Paul R. Schrater,et al.  How Optimal Depth Cue Integration Depends on the Task , 2000, International Journal of Computer Vision.

[18]  A. Yuille,et al.  Bayesian decision theory and psychophysics , 1996 .

[19]  A. Hurlbert,et al.  Perception of three-dimensional shape influences colour perception through mutual illumination , 1999, Nature.

[20]  Judea Pearl,et al.  Probabilistic reasoning in intelligent systems - networks of plausible inference , 1991, Morgan Kaufmann series in representation and reasoning.

[21]  Rajesh P. N. Rao,et al.  Predictive coding in the visual cortex: a functional interpretation of some extra-classical receptive-field effects. , 1999 .

[22]  D H Brainard,et al.  Bayesian color constancy. , 1997, Journal of the Optical Society of America. A, Optics, image science, and vision.

[23]  Edward H. Adelson,et al.  Motion illusions as optimal percepts , 2002, Nature Neuroscience.

[24]  E H Adelson,et al.  Beyond Junctions: Nonlocal form Constraints on Motion Interpretation , 2001, Perception.

[25]  Jean Lorenceau,et al.  Form constraints in motion binding , 2001, Nature Neuroscience.

[26]  James J. Clark,et al.  Data Fusion for Sensory Information Processing Systems , 1990 .

[27]  D. Kersten,et al.  Illusions, perception and Bayes , 2002, Nature Neuroscience.

[28]  Steven K. Feiner,et al.  Computer graphics: principles and practice (2nd ed.) , 1990 .

[29]  Paul Schrater,et al.  Shape perception reduces activity in human primary visual cortex , 2002, Proceedings of the National Academy of Sciences of the United States of America.

[30]  Paul R. Schrater,et al.  Vision, Psychophysics and Bayes , 2001 .

[31]  D. Knill,et al.  The perception of cast shadows , 1998, Trends in Cognitive Sciences.

[32]  Mark S. Drew,et al.  Calculating surface reflectance using a single-bounce model of mutual reflection , 1990, [1990] Proceedings Third International Conference on Computer Vision.

[33]  Karl R. Gegenfurtner.  Visual Perception: Reflections on colour constancy , 1999, Nature.

[34]  Brian E. Smits,et al.  Use of interreflection and shadow for surface contact , 2001, Perception & psychophysics.

[35]  Daniel Kersten,et al.  Inverse 3-D graphics: A metaphor for visual perception , 1997 .

[36]  Song-Chun Zhu,et al.  Minimax Entropy Principle and Its Application to Texture Modeling , 1997, Neural Computation.