We argue that Bayesian decision theory provides a good theoretical framework for visual perception. Such a theory involves a likelihood function specifying how the scene generates the image(s), a prior assumption about the scene, and a decision rule to determine the scene interpretation. This is illustrated by describing Bayesian theories for individual visual cues and showing that perceptual biases found in psychophysical experiments can be interpreted as biases towards prior assumptions made by the visual system. We then describe the implications of this framework for the integration of different cues. We argue that the dependence of cues on prior assumptions means that care must be taken to model these dependencies during integration. This suggests that a number of proposed schemes for cue integration, which only allow weak interaction between cues, are not adequate and instead stronger coupling is often required. These theories require the choice of decision rules and we argue that...
|
1138
|
Robot vision
– Horn
- 1986
|
|
683
|
Statistical Decision Theory and Bayesian Analysis. Corrected reprint of the second
– Berger
- 1993
|
|
345
|
Optimal statistical decisions
– DeGroot
- 1970
|
|
282
|
Planning and control
– Dean, Wellman
- 1991
|
|
200
|
An Essay Toward Solving a Problem in the Doctrine of Chances", Philosophical Transactions of the Royal Society of London 53
– Bayes
- 1995
|
|
193
|
Computational vision and regularization theory
– Poggio, Torre, et al.
- 1985
|
|
115
|
PMF: A stereo correspondence algorithm using a disparity gradient limit
– Pollard, Mayhew, et al.
- 1985
|
|
108
|
Statistical Field Theory
– Parisi
- 1988
|
|
97
|
Data Fusion for Sensory Information Processing Systems
– Clark
- 1990
|
|
80
|
The variational approach to shape from shading
– Horn, Brooks
- 1986
|
|
67
|
Local shading analysis
– Pentland
- 1984
|
|
60
|
Inferring surfaces from images
– Binford
- 1981
|
|
46
|
A parallel algorithm for real-time computation of optical flow
– Bulthoff, Little, et al.
- 1989
|
|
43
|
Analyzing Images of Curved Surface
– Woodham
- 1981
|
|
41
|
An Introduction to Statistical Communication Theory
– Middleton
- 1996
|
|
35
|
Bayes or Bust
– Earman
- 1992
|
|
35
|
Pattern theory: a unifying perspective
– Mumford
- 1996
|
|
33
|
Parallel integration of vision modules
– Poggio, Gamble, et al.
- 1988
|
|
31
|
Tradeoffs between stereopsis and proximity luminance covariance as determinants of perceived 3D structure
– Dosher, Sperling, et al.
- 1986
|
|
30
|
An integrated spatio-temporal approach to automatic visual guidance of autonomous vehicles
– Dickmanns, Mysliwetz, et al.
- 1990
|
|
26
|
Shape from specularities: computation and psychophysics
– BLAKE, BÜLTHOFF
- 1991
|
|
24
|
Minimodularity and the perception of layout
– Bruno, Cutting
- 1988
|
|
23
|
A computational theory for the perception of coherent visual motion
– Yuille, Grzywacz
- 1988
|
|
21
|
The aperture problem. II. Spatial integration of velocity information along contours
– Nakayama, Silverman
- 1988
|
|
19
|
Apparent surface curvature affects lightness perception
– Knill, Kersten
- 1991
|
|
14
|
A Statistical framework for robust fusion of depth information
– Maloney, Landy
- 1989
|
|
13
|
Bayesian method for recovering surface and illuminant properties from photosensor responses
– Brainard, Freeman
- 1994
|
|
13
|
Exploiting the generic view assumption to estimate scene parameters
– Freeman
- 1993
|
|
13
|
The aperture problem. I. Perception of nonrigidity and motion direction in translating sinusoidal lines
– Nakayama, Silverman
- 1988
|
|
13
|
Speech recognition using Hidden Markov Models
– Paul
- 1990
|
|
12
|
Strategy and optimization in human information processing
– Sperling, Dosher
- 1986
|
|
11
|
Interaction of different modules in depth perception
– BUlthoff, Mallot
- 1987
|
|
11
|
Counting every quantum
– Sakitt
- 1972
|
|
10
|
Physical limits to sensation and perception
– Bialek
- 1987
|
|
10
|
Associative learning of scene parameters from images
– Kersten, O'Toole, et al.
- 1987
|
|
10
|
Towards a neural understanding of visual surface representation
– Nakayama, Shimojo
- 1990
|
|
9
|
Classification and clustering in spatial and image data
– Ripley
- 1992
|
|
8
|
Towards a model-based bayesian theory for estimating and recognizing parameterized 3d objects using two or more images taken from different positions
– Cernushi-Frias, Cooper, et al.
- 1989
|
|
8
|
Interaction between transparency and structure from motion
– Kersten, Bülthoff, et al.
- 1992
|
|
6
|
Lightness perception can be affected by surface curvature from stereopsis
– Buckley, Frisby, et al.
- 1994
|
|
6
|
Dependence of speed and direction perception on cinematogram dot density
– Watamaniuk, Grzywacz, et al.
- 1993
|
|
6
|
Stereo Integration, Mean Field Theory and Psychophysics
– Yuille, Geiger, et al.
- 1990
|
|
5
|
Bayesian models for seeing shapes and depth
– Bulthoff
- 1991
|
|
5
|
Integration of stereo, shading and texture
– Bülthoff, A
- 1990
|
|
5
|
Displacement thresholds for coherent apparent motion random dot-patterns
– Ramachandran, Anstis
- 1983
|
|
4
|
Bayesian models, deformable templates and competitive priors
– Yuille, Clark
- 1993
|
|
3
|
Stereopsis and eye movement
– Geiger, Yuille
- 1989
|
|
2
|
An ideal observer model for inference of shape from texture. Vision Research
– Blake, Bulthoff, et al.
- 1993
|
|
2
|
Disparity gradients and depth scaling
– Bulthoff, Fahle, et al.
- 1991
|
|
2
|
A computational theory for the perception of inertial motion
– Grzywacz, Smith, et al.
- 1989
|
