1. Region Filling and Object Removal by Exemplar-Based Image Inpainting

2. Introduction

3. A new algorithm is proposed for removing large objects from digital images.
this problem has been addressed by two classes of algorithms:
1) “texture synthesis” algorithms for generating large image regions from sample textures
2) “inpainting” techniques for filling in small image gaps.

4. Key Observations
A. Exemplar-Based Synthesis Suffices

5. B. Filling Order Is Critical

6. Algorithm
Each pixel maintains a color value (or “empty,” if the pixel is unfilled) and a confidence value.
Algorithm iterates the following three steps until all pixels have been filled.
1) Computing Patch Priorities
2) Propagating Texture and Structure Information
3) Updating Confidence Values

7. the priority computation is biased toward those patches which
1) Computing Patch Priorities
the priority computation is biased toward those patches which
1) are on the continuation of strong edges.
2) are surrounded by high-confidence pixels.
Given a patch centered at the point p for some , we define its priority as the product of two terms

8. C(p) the confidence term that measure of the amount of
reliable information surrounding the pixel p.

9. D(p) the data term that is a function of the strength of isophotes hitting the front at each iteration.
(1) np estimated as the unit vector orthogonal to the line
through the preceding and the successive points in the
list
(2) is computed as the maximum value of the
image gradient in Robust filtering techniques may
also be employed here.

10. 2) Propagating Texture and Structure Information
propagate image texture by direct sampling of the source
region.
the distance between two generic patches and
is simply defined as the sum of squared differences

11. Synthesizing One Pixel
SAMPLE
Infinite sample image
Generated image
Instead of constructing a model, let’s directly search the
input image for all such neighbourhoods to produce a histogram for p

12. Really Synthesizing One Pixel
SAMPLE
finite sample image
Generated image
However, since our sample image is finite, an exact neighbourhood match might not be present
So we find the best match using SSD error (weighted by a Gaussian to emphasize local structure), and take all samples within some distance from that match

13. 3) Updating Confidence Values:
After the patch has been filled with new pixel values,
the confidence is updated in the area delimited by

15. Results And Comparions

16. Time

17. Shape of the select

19. Hand-draw

20. Large object

23. END
THANKS EVERYONE