Slide 1

CS448f: Image Processing For Photography and Vision Blending and Pyramids

Slide 2

Blending Weve aligned our images. What now? Averaging Weighted averaging min/max/median

Slide 3

Noise reduction by Averaging

Slide 4

2 Shots

Slide 5

4 Shots

Slide 6

8 Shots

Slide 7

16 Shots

Slide 8

Noise Reduction by Averaging Were averaging random variables X and Y Both have variance S 2 Variance of X+Y = 2S 2 Std.Dev. of X+Y = sqrt(2). S Std.Dev of (X+Y)/2 = sqrt(2)/2. S Ie, every time we take twice as many photos, we reduce noise by sqrt(2)

Slide 9

Noise Reduction by Averaging Average 4 photos: noise gets reduced 2x Average 8 photos: noise gets reduced 3x Average 16 photos: noise gets reduced 4x

Slide 10

Noise Reduction by Median (demo)

Slide 11

Median v Average

Slide 12

Slide 13

Can we identify the bad pixels? Theyre unlike their neighbours Instead of averaging, weighted average where weight = similarity to neighbours

Slide 14

Weighted Average

Slide 15

Can we identify the bad pixels? Theyre unlike their neighbours Instead of averaging, weighted average where weight = similarity to neighbours Favors blurriness

Slide 16

Input

Slide 17

Other uses of Median Removing Transient Occluders (live demo) (Gates demo) (surf demo)

Slide 18

Panorama Stitching

Slide 19

Slide 20

Slide 21

Slide 22

Slide 23

Multiple Exposure Fusion

Slide 24

Slide 25

Slide 26

Slide 27

Slide 28

Slide 29

Slide 30

Slide 31

Slide 32

Slide 33

Slide 34

Focus Fusion

Slide 35

Slide 36

Slide 37

Slide 38

Slide 39

Pyramids Weve been breaking images into two terms for a variety of apps Coarse + Fine More generally we can break it into many terms: Very coarse + finer + finer... + finest.

Slide 40

Pyramids We can do this by blurring more and more:

Slide 41

Pyramids And then (optionally) taking differences --

Slide 42

Pyramids The coarse layers can be stored at low res. Gaussian Pyramid Laplacian Pyramid

Slide 43

Pyramids How much memory does this use?

Slide 44

Pyramid Uses: Sampling arbitrarily sized Gaussians Equalizing an image The different levels represent different frequency ranges We can scale each frequency level and recombine Blending multiple images

Slide 45

Pyramid Blending Key Insight: Coarse structure should blend very slowly between images (lots of feathering), while fine details should transition more quickly. More robust to tricky cases than plain old compositing

Slide 46

Inputs:

Slide 47

Compositing: Hard Mask

Slide 48

Compositing: Soft Mask

Slide 49

Multi-Band Blending

Slide 50

Exposure Fusion http://research.edm.uhasselt.be/~tmertens/papers/exposure_fusion_reduced.pdf