Motion from Blur

30
Motion from Blur Shengyang Dai and Ying Wu EECS Department, Northwestern University NORTHWESTERN UNIVERSITY

description

NORTHWESTERN UNIVERSITY. Motion from Blur. Shengyang Dai and Ying Wu EECS Department, Northwestern University. Discrete samples over time. Integral over time. Optical flow estimation. Motion from blur. ?. http ://vision.middlebury.edu/flow / - PowerPoint PPT Presentation

Transcript of Motion from Blur

Page 1: Motion from Blur

Motion from BlurShengyang Dai and Ying Wu

EECS Department, Northwestern University

NORTHWESTERNUNIVERSITY

Page 2: Motion from Blur

http://vision.middlebury.edu/flow/Baker, Scharstein, Lewis, Roth,

Black, Szeliski, ICCV’07(courtesy to Simon Baker)

Motion from blur

Optical flow estimation

?

Discrete samples

over time

Integral over time

Page 3: Motion from Blur

LiteratureTask Input Extra info

Q. Shan, W. Xiong, and J. Jia, CVPR 07 Rotational motion blur

Single image

User interaction

A. Levin, NIPS 06

Multiple / local invariant linear motion estimation and

segmentation

Known single blur direction

S. Cho, Y. Matsushita, and S. Lee, ICCV 07 Two images

XL. Bar, B. Berkels, M. Rumpf, and G. Sapiro,

ICCV 07Sequence

Space-variant linear motion estimation from blurred image(s)

Page 4: Motion from Blur

Our workTask Input Extra info

Q. Shan, W. Xiong, and J. Jia, CVPR 07 Rotational motion blur

Single image

User interaction

A. Levin, NIPS 06

Multiple / local invariant linear motion estimation and

segmentation

Known single blur direction

S. Cho, Y. Matsushita, and S. Lee, ICCV 07 Two images

XL. Bar, B. Berkels, M. Rumpf, and G. Sapiro,

ICCV 07Sequence

Our work

1. Global parametric form motion from blur (e.g., affine / rotational motion)

2. Multiple / local motion estima-tion and segmentation from blur

3. Non-parametric motion from blur

Single image X

Space-variant linear motion estimation from blurred image(s)

Page 5: Motion from Blur

)2

()2

(b

pb

pbpb III

+

-b

pb bI

P

pbI

Page 6: Motion from Blur

Motion blur constraint

=( ) ·

)2

()2

(b

pb

pbpb III

Page 7: Motion from Blur

Motion Blur vs. Optic Flow

)2

()2

(b

pb

pbb III

tII m

• Lucas and Kanade, 81• Horn and Schunck, 81• Barron, Fleet, Beauchemin, IJCV 94• Black and Anandan, CVIU 96• Baker and Matthews, IJCV 04• Baker, Scharstein, Lewis, Roth, Black,

Szeliski, ICCV 07• ……

?

Page 8: Motion from Blur

channel image representation

Spectral matting Levin, Rav-Acha, and Lischinski, CVPR 07

Unsupervised image layer decomposition

(courtesy to Anat Levin)0 1

Page 9: Motion from Blur

Alpha motion blur constraint

)2

()2

(b

pb

pbb III )2

()2

(b

pb

pbb 1 bb0

1

b b

0 bb 0 b

• Use to replace the original image

• Assumptions: most pixels have 0 / 1 alpha values

• Observation: mostly when

,b II ,b

0 bwhen

Page 10: Motion from Blur

motion blurred imagebI

1 bb

Alpha motion blur constraint

Page 11: Motion from Blur

1

0

b1 bb

Alpha motion blur constraint

b

Alpha channel of the blurred imageb

Page 12: Motion from Blur

Motion Blur vs. Optic Flow

1 bb )2

()2

(b

pb

pbb III

tII m

• Lucas and Kanade, 81• Horn and Schunck, 81• Barron, Fleet, Beauchemin, IJCV 94• Black and Anandan, CVIU 96• Baker and Matthews, IJCV 04• Baker, Scharstein, Lewis, Roth, Black,

Szeliski, ICCV 07• ……

?

Page 13: Motion from Blur

Hough transform

b

1 bb

Page 14: Motion from Blur

Techniques for optical flow

• Lucas-Kanade• Horn-Schunck• Robust estimation• RANSAC• Multiple resolution

• Parametric• Piecewise smooth• Segmentation• LK meets HS• ……

Page 15: Motion from Blur

Techniques for motion from blur

• Lucas-Kanade• Horn-Schunck• Robust estimation• RANSAC• Multiple resolution

• Parametric• Piecewise smooth• Segmentation• LK meets HS• ……

Page 16: Motion from Blur

Tasks

Page 17: Motion from Blur

Global affine motion blur

Global rotational motion blur

Multiple blur model estimation and segmentation

Non-parametric motion field estimation

Page 18: Motion from Blur

Experiments – affine motion

Page 19: Motion from Blur

Experiments – rotational motion

Page 20: Motion from Blur

Experiments – segmentation

Segmentation result

Page 21: Motion from Blur

Experiments – non-parametric motion

Page 22: Motion from Blur

Applications

Page 23: Motion from Blur

Ground truth image

Synthesized affine blurred image

(PSNR: 21.66dB)

Deblurred result (PSNR: 24.75dB)

Image deblurringModified Richardson-Lucy iteration

average estimation error: 0.43average motion vector: 15.43

Page 24: Motion from Blur

Affine blurred image Our deblur result with• space-variant blur• modified RL iteration

Matlab deconvlucy with • space-invariant blur• original RL iteration

Page 25: Motion from Blur

time Input

Deblurred

Blur synthesis

Motion synthesis

Blur / motion synthesis

Page 26: Motion from Blur

Blur synthesis

Page 27: Motion from Blur

Input Output

Motion synthesis

Page 28: Motion from Blur

Input Output

Motion synthesis

Page 29: Motion from Blur

Summary• Contributions

– A local linear motion blur constraint– Connection between motion blur and optic flow– Space-variant motion estimation from blur– Applications on deblurring and blur/motion synthesis

• Limitations– May not hold for heavily textured region– Rely on robust matting

• Future work– Integrating more algorithms from optic flow

Page 30: Motion from Blur

Motion from BlurShengyang Dai and Ying Wu

EECS Department, Northwestern Universityb

1 bb

Thanks! Questions?