COMP 790-096: Computational Photography
Fall 2008, Tuesdays and Thursdays 12:30-1:45, FB 008
Instructor: Svetlana Lazebnik (lazebnik -at- cs.unc.edu)
Computational photography is an emerging field that aims to overcome the limitations of conventional digital cameras to produce more vivid, compelling, and meaningful visualizations of the world around us. In this course, we will study ways of manipulating and combining photographs and videos to produce new and better pictures, 3D models, walkthroughs, panoramas, animations, etc. Although we will primarily focus on computational techniques that work with standard digital photographs or videos, we will also touch on the subject of designing novel and unconventional imaging devices, such as light field cameras.
A tentative list of topics to be covered includes:
Prerequisites: COMP 665, COMP 776, or basic familiarity with standard techniques of computer graphics and digital image analysis, as well as ability to read and understand computer vision and graphics papers. If you are not sure whether your background is right for the course, don't hesitate to contact me.
If you have taken COMP 790 or COMP 776 with me last year, I really hope to see you in this course! The overlap with the topics covered in those courses will be minimal.
List of Papers (tentative)
Data-Driven Image Synthesis
Removing Camera Shake and Motion Blur
Cutting and Pasting
Image Warping and Morphing
Panoramas, Mosaics, Collages
Single- and Multiple-View Reconstruction
Beyond Conventional Cameras
Digital Image Forensics
Additional Papers (good for project ideas)
A. Oliva, A. Torralba, and P.G. Schyns
L. Zelnik-Manor and P. Perona
5th International Symposium on Non-Photorealistic Animation and Rendering, 2007
J. Sivic, B. Kaneva, A. Torralba, S. Avidan, and W. Freeman
Internet Vision Workshop, 2008
Y. Liu, T. Belkina, J. Hays, and R. Lublinerman