Title: Computational photography MIT 6.098, 6.882 Bill Freeman, Fredo Durand
1Computational photography MIT 6.098, 6.882 Bill
Freeman, Fredo Durand
- Finish digital forensics
- Analyzing multiple images
- Shapetime photography
- Image stacks
- Analysing and synthesizing motion sequences
- Motion without movement
- Motion magnification
April 20, 2006
2Analyzing multiple images
- Bill Freeman
- Fredo Durand
- MIT Computational Photography, 6.882
- April 20, 2006
3Multiple-exposure images by Marey
4Strobe photograph by Edgerton
5Other photographs by Doc Edgerton
6What hardware was needed to make these
photographs?
- Strobe light, capacitor, thyristor
7Computational photography
- Surely we can update those photographic
techniques, adding the generality and flexibility
of digital methods. Analyze and re-render the
images.
8Computational photography
- Fredo and Bill describing computational
photography - Fredo using computation to make better quality
photographsto enhance. - Bill using computation to reveal things about
the world that we otherwise couldnt seeto
reveal.
9- How display a single-frame summary of multiple
frames?
10(No Transcript)
11Typical frame
12Average over 50 frames
13Median filter over time
14Vector median filter (20x20 patchs)
15- 2x2 vector median 2x2 vector least median
16Shapetime photography
- Joint work with Hao Zhang, U.C. Berkeley
- 2002
17Video frames
18(No Transcript)
19Resulting composite image
Frame index of each displayed pixel
20Frame index of each displayed pixel
Resulting composite image
With edge-preserving regularization
21(No Transcript)
22(No Transcript)
23(No Transcript)
24(No Transcript)
25Z-cam, made by 3DV
http//www.3dvsystems.com
263DV camera operation
http//www.3dvsystems.com
273DV camera operation
http//www.3dvsystems.comv
283DV camera operation
http//www.3dvsystems.com
29RGB image
30Z image
31shapetime video image
32(No Transcript)
33Zitnick et al, Siggraph 2004
34- Show Michael Cohen slides, a selection from
- http//research.microsoft.com/cohen/FindingMagic
InAnImageStack.pdf
35(No Transcript)
36(No Transcript)
37(No Transcript)
38(No Transcript)
39(No Transcript)
40(No Transcript)
41(No Transcript)
42(No Transcript)
43(No Transcript)
44- Demonstrate MSR group shot program, downloadable
from - http//research.microsoft.com/cohen/
- or
- http//research.microsoft.com/projects/GroupShot/
45Analyzing and synthesizing motion
- Bill Freeman
- Fredo Durand
- MIT Computational Photography, 6.882
- April 20, 2006
46Aperture Problem and Normal Flow
47Aperture Problem and Normal Flow
48Aperture Problem and Normal Flow
49Aperture Problem and Normal Flow
50Aperture Problem and Normal Flow
51Aperture Problem and Normal Flow
52Optical flow constraint equation
53Aperture Problem and Normal Flow
The gradient constraint
Defines a line in the (u,v) space
Normal Flow
54Combining Local Constraints
v
etc.
u
55Lucas-Kanade (a good, generic motion analysis
method) Integrate gradients over a patch
Assume a single velocity, u, v, for all pixels
within an image patch. Find the (u, v) that
minimizes the BCCE squared residual over the
patch
Setting derivative w.r.t. (u, v) equal to zero
gives
Note similarity of LHS matrix to Harris corner
detector. When full-rank (corner-like),
specifies a unique (u, v).
56Motion without movement
- Joint work with Ted Adelson and David Heeger, MIT
- 1991
57A linear combination of quadrature-phase filters
can advance the local phase
58Convolved with an image, the image data now
modulates the local amplitude. People
mis-attribute the phase advance to translation.
(Steerable filters allow synthesizing motion in
arbitrary directions.)
59Motion without movement video
60http//www.cs.yorku.ca/kosta/Motion_Without_Movem
ent/Motion_Without_Movement.html
61http//www.cs.yorku.ca/kosta/Motion_Without_Movem
ent/Motion_Without_Movement.html
Konstantinos G. Derpanis
62Motion Magnification