Top Banner
MIDAG@UNC MIDAG@UNC Geometry-driven Geometry-driven Diffusion Diffusion
16
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: MIDAG@UNC Geometry-driven Diffusion. MIDAG@UNC MIDAG@UNC.

MIDAG@UNCMIDAG@UNC

Geometry-driven Geometry-driven DiffusionDiffusion

Geometry-driven Geometry-driven DiffusionDiffusion

Page 2: MIDAG@UNC Geometry-driven Diffusion. MIDAG@UNC MIDAG@UNC.

MIDAG@UNCMIDAG@UNC

Geometry-driven Geometry-driven DiffusionDiffusion

Geometry-driven Geometry-driven DiffusionDiffusion

Page 3: MIDAG@UNC Geometry-driven Diffusion. MIDAG@UNC MIDAG@UNC.

MIDAG@UNCMIDAG@UNC

Geometry-driven Geometry-driven DiffusionDiffusion

Geometry-driven Geometry-driven DiffusionDiffusion

Page 4: MIDAG@UNC Geometry-driven Diffusion. MIDAG@UNC MIDAG@UNC.

MIDAG@UNCMIDAG@UNC

Geometry-driven Geometry-driven DiffusionDiffusion

Geometry-driven Geometry-driven DiffusionDiffusion

Page 5: MIDAG@UNC Geometry-driven Diffusion. MIDAG@UNC MIDAG@UNC.

MIDAG@UNCMIDAG@UNC

Geometry-driven Geometry-driven DiffusionDiffusion

Geometry-driven Geometry-driven DiffusionDiffusion

Page 6: MIDAG@UNC Geometry-driven Diffusion. MIDAG@UNC MIDAG@UNC.

MIDAG@UNCMIDAG@UNC

Geometry-driven Geometry-driven DiffusionDiffusion

Geometry-driven Geometry-driven DiffusionDiffusion

Page 7: MIDAG@UNC Geometry-driven Diffusion. MIDAG@UNC MIDAG@UNC.

MIDAG@UNCMIDAG@UNC

Geometry-driven Geometry-driven DiffusionDiffusion

Geometry-driven Geometry-driven DiffusionDiffusion

Page 8: MIDAG@UNC Geometry-driven Diffusion. MIDAG@UNC MIDAG@UNC.

MIDAG@UNCMIDAG@UNC

Geometry-driven Geometry-driven DiffusionDiffusion

Geometry-driven Geometry-driven DiffusionDiffusion

Page 9: MIDAG@UNC Geometry-driven Diffusion. MIDAG@UNC MIDAG@UNC.

MIDAG@UNCMIDAG@UNC

Geometry-driven Geometry-driven DiffusionDiffusion

Geometry-driven Geometry-driven DiffusionDiffusion

Page 10: MIDAG@UNC Geometry-driven Diffusion. MIDAG@UNC MIDAG@UNC.

MIDAG@UNCMIDAG@UNC

Geometry-driven Geometry-driven DiffusionDiffusion

Geometry-driven Geometry-driven DiffusionDiffusion

Page 11: MIDAG@UNC Geometry-driven Diffusion. MIDAG@UNC MIDAG@UNC.

MIDAG@UNCMIDAG@UNC

Geometry-driven Geometry-driven DiffusionDiffusion

Geometry-driven Geometry-driven DiffusionDiffusion

Page 12: MIDAG@UNC Geometry-driven Diffusion. MIDAG@UNC MIDAG@UNC.

MIDAG@UNCMIDAG@UNC

Geometry-driven Geometry-driven DiffusionDiffusion

Geometry-driven Geometry-driven DiffusionDiffusion

Page 13: MIDAG@UNC Geometry-driven Diffusion. MIDAG@UNC MIDAG@UNC.

MIDAG@UNCMIDAG@UNC

Linear Linear Diffusion Diffusion Results in Results in Drifting Drifting MaximaMaxima

Linear Linear Diffusion Diffusion Results in Results in Drifting Drifting MaximaMaxima

Page 14: MIDAG@UNC Geometry-driven Diffusion. MIDAG@UNC MIDAG@UNC.

MIDAG@UNCMIDAG@UNC

Perona Perona Diffusion Diffusion

Keeps Keeps Some Some Edges Edges

Through Through Scale Scale

Perona Perona Diffusion Diffusion

Keeps Keeps Some Some Edges Edges

Through Through Scale Scale

Page 15: MIDAG@UNC Geometry-driven Diffusion. MIDAG@UNC MIDAG@UNC.

MIDAG@UNCMIDAG@UNC

Compare MargarineCompare Margarine Compare MargarineCompare Margarine

Page 16: MIDAG@UNC Geometry-driven Diffusion. MIDAG@UNC MIDAG@UNC.

MIDAG@UNCMIDAG@UNC

Bilateral Filtering: VCD and form Bilateral Filtering: VCD and form systemsystem

Bilateral Filtering: VCD and form Bilateral Filtering: VCD and form systemsystem

Think: form system input to Think: form system input to diffusion process, and diffusion.diffusion process, and diffusion.

Blurs within boundary by using a Blurs within boundary by using a weighting function that is the weighting function that is the product of two Gaussians:product of two Gaussians: Gaussian with a spatial kernel: Gaussian with a spatial kernel:

closer pixels have higher weight. closer pixels have higher weight. Gaussian in the intensity domain: Gaussian in the intensity domain:

higher weights for pixels with similar higher weights for pixels with similar intensities.intensities.

Dorsey, 2002Dorsey, 2002

Think: form system input to Think: form system input to diffusion process, and diffusion.diffusion process, and diffusion.

Blurs within boundary by using a Blurs within boundary by using a weighting function that is the weighting function that is the product of two Gaussians:product of two Gaussians: Gaussian with a spatial kernel: Gaussian with a spatial kernel:

closer pixels have higher weight. closer pixels have higher weight. Gaussian in the intensity domain: Gaussian in the intensity domain:

higher weights for pixels with similar higher weights for pixels with similar intensities.intensities.

Dorsey, 2002Dorsey, 2002