Image Modeling - continued

1
Image Modeling - continued

• Given observed feature statistics H(a)obs, we
  associate an energy with any image I as
• Then the corresponding Gibbs distribution is
• The q(I) can be sampled using a Gibbs sampler or
  other Markov chain Monte-Carlo algorithms

2
Image Modeling - continued

• Image Synthesis Algorithm
• Compute Hobs from an observed texture image
• Initialize Isyn as any image, and T as T0
• Repeat
• Randomly pick a pixel v in Isyn
• Calculate the conditional probability q(Isyn(v)
  Isyn(-v))
• Choose new Isyn(v) under q(Isyn(v) Isyn(-v))
• Reduce T gradually
• Until E(I) lt e

3
A Texture Synthesis Example
Observed image
Initial synthesized image
4
A Texture Synthesis Example

• Energy and conditional probability of the marked
  pixel

5
A Texture Synthesis Example - continued

• A white noise image was transformed to a
  perceptually similar texture by matching the
  spectral histogram

6
A Texture Synthesis Example - continued

• Synthesized images from different initial
  conditions

7
Texture Synthesis Examples - continued

• A random texture image

8
Texture Synthesis Examples - continued
Observed image
Synthesized image

• An image with periodic structures

9
Texture Synthesis Examples - continued
Mud image
Synthesized image

• A mud image with some animal foot prints

10
Texture Synthesis Examples - continued
Observed image
Synthesized image

• A random texture image with elements

11
Texture Synthesis Examples - continued
Observed image
Synthesized image

• An image consisting of two regions
• Note that wrap-around boundary conditions were
  used

12
Texture Synthesis Examples - continued

• A cheetah skin image

13
Texture Synthesis Examples - continued
Observed image
Synthesized image

• An image consisting of circles

14
Texture Synthesis Examples - continued
Observed image
Synthesized image

• An image consisting of crosses

15
Texture Synthesis Examples - continued

• A pattern with long-range structures