Vision and Robotics Seminar
School of Computer Science and Engineering
The Hebrew University
will speak on
Multi-Sensor Super Resolution
Image sensing is usually done with multiple sensors, like the RGB sensors in color imaging and the IR and EO sensors in surveillance. A method to increase the resolution of each sensor by considering the other sensors is presented, based on the super-resolution approach originally used with images of the same sensor. The multi-sensor extension is enabled by an assumption of local affine relations between the intensities of different sensors at corresponding image locations.
A benefit of Multi-sensor super-resolution is the ability to increase
the resolution of even a single color image (e.g. increase the ``red" resolution
based on the ``green" and ``blue" images). Since the displacement
between different color sensors is usually fixed in digital color cameras,
and since the different sensors are capturing the images simultaneously,
it allows the same computation for all pictures taken with this camera,
even in dynamic scenes.
The lecture will take place in the
Lecture Hall, Room 1, Ziskind Building
on Thursday, July 4, 2002