|
Statistical Properties of Natural Images
|
|
A "prior" probability model for visual images (i.e., a specification
of the likelihood that any given image would be seen) provides a
powerful constraint for many applications in image processing,
computer vision, and graphics. In addition, it seems likely that the
statistical properties of images are fundamental in shaping the design
of biological visual systems through evolution, development, learning
and adaptation. We've studied statistical image properties, developed
developed parametric models for these properties, and used these
models in a variety of applications, and as a basis for understanding
the representations used by mammalian visual systems:
|
 |
Statistical modeling:
nips-06a,
Bovik-05a,
JMIV-03,
ACHA-00,
spie-00b,
nips-99,
spie-99,
asil-97
|
|
Relationship to neurobiology (the "efficient coding" hypothesis):
CON-03,
Rao-02,
NN-01,
AnnRev-01,
nips-00,
arvo-99a,
arvo-99b,
nips-98,
arvo-98
|
|
Characterization of neurons with stochastic stimuli:
NN-06,
JOV-06,
JN-05,
Neuron-05,
vss-05,
NC-04,
cns-04,
Gazzaniga-04,
NW-03,
nips-03b,
cns-03a,
TICS-03,
cns-02,
nips-01,
vss-01
|
|
Perceptual Image Metrics:
IP-06,
icip-05,
Bovik-05b,
icassp-05,
spie-05,
nips-04,
spie-04,
asilomar-03,
IP-04a
|
|
Denoising:
icip-07a,
nips-06b,
nips-06a,
IP-03,
icip-03,
TR-02,
icip-01,
icip-00a,
icip-00b,
spie-00a,
Vidakovic-99,
icip-96a
|
|
Enhancement:
icip-05,
nips-03a
|
|
Texture analysis/representation/synthesis:
IJCV-00,
stv-99,
icip-98
|
|
Compression:
IP-06,
IP-99,
icip-97,
icassp-97,
MS-thesis-88
|
|
Visual Motion: Analysis and Representation
|
|
When we move, the visual images projected onto our retinae change
accordingly. For both biological and computer vision systems, the
pattern of image velocities (sometimes called optic flow) carries
important environmental information. We've studied this problem from
a variety of different angles, emphasizing the issues and constraints
that are common to all:
|
|
General:
PhD-thesis-93
|
|
Modeling - perception:
NN-06,
nips-05,
cosyne-05,
nips-04,
NN-02,
Nature-01,
NN-00,
VR-98b,
Harris-94,
arvo-94b,
arvo-92a
|
|
Modeling - physiology:
NN-06,
Chalupa-03,
sfn-02,
vss-02,
NN-01a,
VR-98a,
arvo-96,
pnas-96,
arvo-94a
arvo-93a
|
|
Estimation of optic flow:
Jaehne-99,
mdsp-93,
TR-202-92,
cvpr-91,
arvo-91
|
|
Estimation of depth/heading:
caip-97b,
Landy-96,
mw-91
|
|
Motion estimation/segmentation:
eccv-94,
cvpr-93,
arvo-93b,
arvo-92b
|
|
Image Analysis and Representation
|
|
The features that occur in visual images are often oriented (e.g.,
contours), and often have a particular size or scale. A wide variety of
multi-scale, multi-orientation representations have been developed over
the past few decades, and have proven to be important for solving
problems in image processing and computer vision:
|
|
Nonlinear multi-scale representation:
spie-07,
IP-06,
vlbv-03.
|
|
Orientation analysis / derivative filter design:
TR-05,
IP-04b,
caip-97a,
IP-96,
icip-96b,
icip-94
|
|
Multi-scale, oriented representations (steerable pyramids):
icip-95,
IT-92
|
|
Orthogonal multi-scale image representations (wavelets):
Woods-90,
IEEE-89,
MS-thesis-88,
spie-87
|
|
Range estimation:
JOSA-98,
eccv-96
|
|
Rendering using illumination bases:
TR-97,
Presence-95,
egw-94
|
