Download presentation
Presentation is loading. Please wait.
Published byChloe Howe Modified over 10 years ago
1
Inferring the kernel: multiscale method Input image Loop over scales Variational Bayes Upsample estimates Use multi-scale approach to avoid local minima: Initialize 3x3 blur kernel Convert to grayscale Remove gamma correction User selects patch from image
2
Image Reconstruction Input image Full resolution blur estimate Non-blind deconvolution (Richardson-Lucy) Deblurred image Loop over scales Variational Bayes Upsample estimates Initialize 3x3 blur kernel Convert to grayscale Remove gamma correction User selects patch from image
4
Results on real images Submitted by people from their own photo collections Type of camera unknown Output does contain artifacts – Increased noise – Ringing Compares well to existing methods
5
Original photograph
6
Blur kernel Our output
7
Original photograph Matlabs deconvblind
8
Original Our output Close-up of garland Matlabs deconvblind
9
A submitted photograph A small list of the reasons why we didnt attempt this photograph: Most of the features of interest are saturated. Different blur kernels for different lights (compare lantern streaks with sky light streaks and different than the water reflection streaks, and the car streaks below bridge and the streaks to left of bridge.) The objects reflected by flash are stationary and have no motion blur.
10
We dont handle subject motion blur
11
Failure mode
12
Original photograph
13
Matlabs deconvblind
14
Photoshop sharpen more
15
Our output Blur kernel
17
Original photograph
18
Our output Blur kernel
19
Original photograph
20
Our output Blur kernel
21
Matlabs deconvblind
22
Original photograph
Similar presentations
© 2024 SlidePlayer.com. Inc.
All rights reserved.