Hybrid Camera for High-Resolution Software Refocus and Improved Depth of Field
In comparison to conventional digital cameras, current light field cameras are still limited in spatial resolution and depth of field control. Using a new camera combination, Vivek Boominathan and colleagues from Rice University were able to lift these limitations.
In a recent publication, the authors present a prototype hybrid imaging system consisting of a Lytro light field camera (380 x 380 pixel spatial resolution) and a conventional Canon DSLR (18 megapixel resolution). The setup doesn’t require co-location or prior calibration of the two cameras. In conjunction with a “simple patch-based algorithm”, the researchers were able to produce a high-resolution light field with 9 times the Lytro camera’s resolution and about 1/9th of the camera’s depth of field.
The research will be presented at the IEEE International Conference on Computational Photography 2014, which takes place in Santa Clara on May 2-4, 2014: Vivek Boominathan, Kaushik Mitra, Ashok Veeraraghavan, “Improving Resolution and Depth-of-Field of Light Field Cameras Using a Hybrid Imaging System”, IEEE International Conference on Computational Photography (ICCP), 2014.
Actually, I have applied a patent about this kind of hybrid framework for high resolution lightfield more than a year ago. Some experimental results had gotten by using the similar methods.
Project webpage: http://www.ece.rice.edu/~vb10/projects/2014/HybridHighResLF/HybridHighResLF.html