Skip to content

Tom-Zheng/exemplar-based-depth

Repository files navigation

CS7495 Final Project
Sooraj Bhat

Object Removal by Exemplar-based Inpainting


USING THE CODE

Everything was done in Matlab and MEX (i.e. a C function callable from
Matlab).  First, the C code needs to be compiled.  At the Matlab
prompt, type:

   mex bestexemplarhelper.c

You should then be able to run the following (which takes ~2mins on my
2GHz Pentium 4 laptop with 256MB of RAM):

   [i1,i2,i3,c,d,mov] = inpaint('bungee0.png','bungee1.png',[0 255 0]);

A smaller run (~25 seconds) would be:

   [i1,i2,i3,c,d,mov] = inpaint('bw0.png','bw2.png',[0 255 0]);

To visualize:

   plotall;
   close; movie(mov);


SAMPLE IMAGES & RESULTS

*0.png	      The original images ('bw' and 'bungee')
*{1,2}.png    The different fill region masks.
*P2.png	      Plots of confidence and data terms.
*P5.png	      Plots of original, fill region, inpainted images and
              confidence and data terms. 

Other images in the dir contain redundant information.  I didn't
include the movies since they are big, but they are available on my
project page (AVI format).

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published