CVPR 2010 Tutorial
Monday, June 14, 2010 (Half Day, AM Session)
Amit Agrawal,
Ashok Veeraraghavan, Srinivasa Narasimhan and Ankit Mohan
Computational imaging
offers imaging functionalities beyond
traditional imaging. Recently, we have seen tremendous interest in new
forms of
sampling the visual world. These include capturing the angular
information in a
light field camera for view interpolation and digital refocusing,
modulating
the temporal integration pattern of a camera for motion deblurring via
coded
exposure, modulating the aperture pattern for depth estimation and
digital
refocusing etc. Coupled with powerful reconstruction algorithms,
computational
imaging simplifies several vision problems.
In this course, we will
present an extensive overview of
computational imaging techniques while providing key ideas and insights
into
their workings. The participants will learn about topics related to
computer
vision, computational photography and computer graphics. We hope to
provide
enough fundamentals to satisfy the technical specialist as well as
tools/softwares
to aid graphics and vision researchers, including graduate students.
(a)
History
of
images
and
imaging
(b) Introduction to Computational Imaging
(c) Course details
Assorted Pixels (Srinivas, 20 mins) pdf
(a) Dimensions of imaging
(b) Multiplexing and Multi-dimensional sampling
(c) Examples of assorted pixel cameras
(intensity, color,
polarization, depth, focus, angle, time, etc).
(d) HDR Imaging and Videography.
Coding and
Modulation in Cameras (Amit, 45 mins) pdf
(a) Photo manipulations versus
Coded Imaging
(b) Key concepts: PSF
Invertibility and PSF Invariance
(c) Techniques for Handling Motion
Blur and Defocus Blur
(d) Strobing Camera: High Speed
Imaging for Periodic Signals using Compressive Sensing
(e) Coded Sampling for Temporal
Super-Resolution
Light Fields
and Applications (Ankit, 60 mins) pdf
(a) Introduction to Light Fields
(b) Light Field Capture Methods
(c) Shield Fields
(d) Glare Reduction
(e) Re-Interpretable Imaging: Varying
Resolution in Post-Capture
(f) Bi-Di Screen
for 3-D Gesture Interaction
(g) Bokode: Camera based Interaction from a
Distance
Computational Illumination (Srinivas, 45 mins) pdf
(a) Geometry Acquisition
(b) Acquiring Material
Properties and Appearance
(c) Acquiring Volumetric
Media
(d) High Speed Vision
(e) Novel Displays
Future Directions (Amit, 15 mins) pdf
(a) Imaging beyond Computer Vision
(b) Smart Optics, Smart Sensors
and Smart Lighting
(c) New Challenges in Image
Authentication, Privacy and Social Impact