EE Seminar: FPGA platform for real-time image enhancement based on sparse coding
Speaker: Shachar Yoseff,
M.Sc. student under the supervision of Prof. Alex Bronstein
Wednesday, March 22nd, 2017 at 15:00
Room 011, Kitot Hashmal., Faculty of Engineering
FPGA platform for real-time image enhancement based on sparse coding
Abstract
We present a proof-of-concept end-to-end FPGA system for real-time image enhancement based on sparse coding. The system uses a fast non-iterative reconstruction algorithm operating with a constant latency in fixed-point arithmetic and achieving real-time performance in a prototype FPGA implementation. We demonstrate three different applications implemented using this generic system: demosaicing, deblurring and extended depth of field imaging. The output of the system, evaluated on simulated and real-life scenes, is qualitatively and quantitatively better than or is comparable to the result of state-of-the-art algorithms.