26
Image Inpainting M. Bertalmio, A.L. Sapiro, V. Caselles, C. Ballester M. Bertalmio, A.L. Bertozzi & G. Sapiro presented by Roman Stanchak Navier Stokes, Fluid Dynamics and Image and Video Inpainting Navier Stokes, Fluid Dynamics and Image and Video Inpainting

Image Inpainting

Embed Size (px)

DESCRIPTION

M. Bertalmio, A.L. Sapiro, V. Caselles, C. Ballester M. Bertalmio, A.L. Bertozzi & G. Sapiro presented by Roman Stanchak. Image Inpainting. Navier Stokes, Fluid Dynamics and Image and Video Inpainting. Navier Stokes, Fluid Dynamics and Image and Video Inpainting. Goal. - PowerPoint PPT Presentation

Citation preview

Page 1: Image Inpainting

Image InpaintingM. Bertalmio, A.L. Sapiro, V. Caselles,

C. Ballester

M. Bertalmio, A.L. Bertozzi & G. Sapiro

presented byRoman Stanchak

Navier Stokes, Fluid Dynamics and Image and

Video Inpainting

Navier Stokes, Fluid Dynamics and Image and

Video Inpainting

Page 2: Image Inpainting

Goal

● Given image with significant portions missing or damaged

● Reconstitute missing regions with data consistent with the rest of the image

Page 3: Image Inpainting
Page 4: Image Inpainting
Page 5: Image Inpainting

Cracks in Photos

Page 6: Image Inpainting

Scratches in Film

Page 7: Image Inpainting

Add or remove elements

Page 8: Image Inpainting

Professional Inpainters Approach

● Goal: – Subjectively restore unity to artwork

● General Approach:– Structure of surrounding area continued

into gap– Color of surrounding area is continued

into gap– Texture is added

Page 9: Image Inpainting

Gestalt Law of Continuation

● Principle holding that there is an innate tendence to perceive a line as continuing its established direction

Page 10: Image Inpainting

General Algorithm

● Iteratively– Continue structure of surrounding area

into gap– Continue color of surrounding area into

gap

Page 11: Image Inpainting

Algorithm

● Define image update as follows:● It(i,j) = dL(i,j) . N(i,j)

– It(i,j) is change in intensity to used to update the image

– L(i,j) is the information to propagate – N(i,j) is the propagation direction– dL(i,j) . N(i,j) is the change in information

along the propagation direction

Page 12: Image Inpainting

Propagation Direction

● Want to preserve angle of 'arrival'● Propagate along isophotes

– lines of equivalent intensity levels

Isophotes

Page 13: Image Inpainting

Propagation Direction

● More formally:– Image gradient

● largest change– Perpindicular

● smallest change● this is the isophote

Isophote is this boundary

= N(i,j)

Page 14: Image Inpainting

Algorithm

● What information to propagate?– 'smoothness'

● Laplacian == Ixx(i,j) +Iyy(i,j) = L(i,j)

Page 15: Image Inpainting

Algorithm

● Repeat this image update step to restore image

● When to stop?– Stop when update value is negligible

● i.e. It(i,j) < epsilon for all i,j– Stop after n iterations

Page 16: Image Inpainting

Some results

Page 17: Image Inpainting

Some results

Page 18: Image Inpainting

Some results

Page 19: Image Inpainting

More results

Page 20: Image Inpainting

More results

Page 21: Image Inpainting

More results

Page 22: Image Inpainting

Navier-Stokes equations

● Nonlinear partial differential equations● Describe the flow of fluids such as

liquids and gases– Air currents– Ocean currents– Water flow in a pipe

Page 23: Image Inpainting

Navier-Stokes equations

Page 24: Image Inpainting

Relation to Inpainting

● Inpainting– Propagation of intensity along isophote

according to smoothness● Fluid mechanics

– Convection of fluid along velocity field according to vorticity

Page 25: Image Inpainting

Relation to Inpainting

● Mapping– Image intensity == Stream function– Isophote direction == fluid velocity– Smoothness == vorticity

● Propels inpainting into an established field with a wealth of theoretical and numerical literature!

Page 26: Image Inpainting

Super Resolution Result

Up scaled image Bicubic sampled Navier-Stokes inpainting