Synthetic and real data set for compressive HDR light field imaging. There are 85 real light fields that are captured with Lytro camera with 5 exposures. The LDR light fields are decoded by [1] and merged view-per-view into an HDR light field using [2].
The synthetic data set is produced using Blender and contains 5 different scenes. Each scene was rendered using a 7x7 grid of cameras. Moreover, we have also produced different baselines, i.e. the distance between the cameras in the 7x7 grid.
For more infomation and questions please contact Saghi Hajisharif : saghi.hajisharif at liu.se or Ehsan Miandji ehsan.miandji at liu.se.
[1] Dansereau, Donald G., Oscar Pizarro, and Stefan B. Williams. "Decoding, calibration and rectification for lenselet-based plenoptic cameras." Proceedings of the IEEE conference on computer vision and pattern recognition. 2013.
[2] Debevec, Paul E., and Jitendra Malik. "Recovering high dynamic range radiance maps from photographs." ACM SIGGRAPH 2008 classes. 2008. 1-10.
If you found the dataset or our published paper useful please consider citing:
