Robust Patch-Based HDR Reconstruction of Dynamic Scenes


Proceedings of SIGGRAPH Asia 2012

ACM Transactions on Graphics (TOG)
Vol. 31, No. 6, Art. 203, November 2012

     Pradeep Sen1          Nima Khademi Kalantari1          Maziar Yaesoubi2          Soheil Darabi2     
     Dan B Goldman3          Eli Shechtman3     


1 University of California, Santa Barbara         2 UNM Advanced Graphics Lab         3 Adobe      

Abstract

High dynamic range (HDR) imaging from a set of sequential exposures is an easy way to capture high-quality images of static scenes, but suffers from artifacts for scenes with significant motion. In this paper, we propose a new approach to HDR reconstruction that draws information from all the exposures but is more robust to camera/scene motion than previous techniques. Our algorithm is based on a novel patch-based energy-minimization formulation that integrates alignment and reconstruction in a joint optimization through an equation we call the HDR image synthesis equation. This allows us to produce an HDR result that is aligned to one of the exposures yet contains information from all of them. We present results that show considerable improvement over previous approaches.


Paper and Additional Materials




Additional data sets with RAW and JPEG image pairs


We also share additional data sets which contain RAW and JPEG images for different scenes. Each of the following zip files contains 7-8 scenes and each scene folder contains both RAW and JPEG images. The exposure values can be computed by using EXIF data attached to each JPEG file. Please cite our paper if you use the following data sets.


Bibtex

@article{PatchBasedHDR,

author = {Pradeep Sen and Nima Khademi Kalantari and Maziar Yaesoubi and Soheil Darabi

and Dan B Goldman and Eli Shechtman},

title = {{Robust Patch-Based HDR Reconstruction of Dynamic Scenes}},

journal = {ACM Transactions on Graphics (TOG) (Proceedings of SIGGRAPH Asia 2012)},

volume = {31},

number = {6},

articleno = {203},

pages = {203:1--203:11},

year = {2012},

}





The information (source code, data sets, etc.) provided on this website is for non-commercial, research purposes only. For commercial inquiries regarding the HDR reconstruction algorithm presented in this paper, please contact Ms. Jovan Heusser at the UNM technology transfer office STC.UNM. Her telephone number is (505) 272-7908 and email address is jheusser@stc.unm.edu.

This page contains documents, source code, videos and other files that could be protected by copyright. They are provided here for reasonable academic fair use. Interested parties are referred to the official published version of the documents which are available from the copyright holder through the external link.