1. Computational Photography
Prof. Feng Liu
Spring 2017
04/03/2017
Computer Graphics

2. Today
Course overview
Admin. Info
Computational Photography

3. People Lecturer: Prof. Feng Liu Room FAB 120-09
Office Hours: MW 3:30-4:30pm

4. Web and Computer Account
Course website
Class mailing list
Everyone needs a Computer Science department computer account
Get account at CAT

5. Textbooks & Readings Computer Vision: Algorithms and Applications
By R. Szeliski
Available online, free
Learning OpenCV 3: Computer Vision in C++ with the OpenCV Library
By Adrian Kaehler and Gary Bradski
Or its early version
Learning OpenCV: Computer Vision with the OpenCV Library
By Gary Bradski and Adrian Kaehler
Papers recommended by the lecturers

6. Grading
40%: Readings
60%: Project

7. Readings About 2 papers every week
Write a brief summary for one of the papers
Totally less than 500 words
What problem is addressed?
How it is solved?
The advantages of the presented method?
The limitations of the presented method?
How to improve this method?

8. Project Literature Study Option System Option Research Option
OK, not exciting…
Read a rich set of literature on one topic
And write a survey paper
System Option
Good and popular
Implement an easy-to-use system based on an existing algorithm
Research Option
Excellent, less popular
Define a new problem and solve it
Or, develop a new solution to an existing problem

9. Project
Group options
1 or 2 members for the System and Research Options
Credits will evenly divided among group members
Only 1 member for the Literature Study Option

10. Project Timeline 04/17: Project proposal due
Submit a short project proposal
300 to 500 words
06/05-06/07: In-class project presentation
Around 10 minutes
5pm, 06/09: Final report due
Submit a report, test data set and source code
Late submission policy
Your project will be accepted until 5pm, 06/09
But, will be penalized according to G=G0*(1-n*0.05/24), where n is the number of hours delayed, G0 is the raw score, and G is your final score.

11. Programming tools C/C++ under Windows Matlab Others
Preferable Visual Studio 2008, 2010, or 2012
Highly recommended
You can use the OpenCV libraries
Other graphics and vision libraries
Matlab
Recommended
Others OK
As long as it works for you

12. OpenCV Open Source Computer Vision library http://opencv.org/ V 3.2
Perhaps the most popular toolkits for computer vision
Provides APIs for a wide range of vision algorithms
Highly recommended for your project

13. Lab Facilities EB 325: Windows Lab
Visual Studio 2012 installed
You need a Computer Science department account to use any CS labs
Request one from CAT

14. Admin Questions?

15. Today
Course overview
Admin. Info
Computational Photography

16. What Is Computational Photography
An extension of traditional (digital) photography that combines computational techniques from computer vision and computer graphics for improving image making
Computer Graphics
Computer Vision CP
Digital Photography and Optics
Slide credit: C. Dyer

17. Filter: De-noise noisy image naïve denoising Gaussian blur
The higher the ISO number, the faster the camera sensor absorbs light.
An increase in ISO also causes an increase in image noise.
noisy image
naïve denoising Gaussian blur
better denoising edge-preserving filter
Slide credit: Sylvain Paris and Frédo Durand

18. De-Blur
Original photograph Output
Slide credit: Fergus et al.

19. Super-resolution
Reprint from Fattal 07

20. Super-resolution
Reprint from Liu et al. 2001

21. Color2Gray
New Algorithm
Color
Grayscale
Slide credit: Gooch et al.

22. Tone Adjustment Input Output Reprint from Lischinski et al. 06

23. Shadow Editing
Input Output

24. High Dynamic Range Imaging
Reprint from G. Brown 2006

25. Panorama
Reprint from Brown and Lowe 2007

26. Segmentation & Matting
Photos from:

27. Picturing Place: Building Rome in a Day
Input Output
Photo from:

28. Picturing Place: Photo Tourism
Photo from:

29. Stabilization

30. Stereoscopic 3D Image source:

31. Stereoscopic 3D

32. Stereoscopic 3D
Image source:

33. Computational Stereoscopic Cinematography

34. Computational Stereoscopic Cinematography

35. Next Time
Camera