Accelerate Local Tone Mapping for High Dynamic Range Images Using OpenCL with GPU
Department of Electrical Engineering, National Tsing Hua University, Hsinchu, Taiwan 30013
The 23rd Int’l Conf on Parallel and Distributed Processing Techniques and Applications (PDPTA’17), 2017
@article{liao2017accelerate,
title={Accelerate Local Tone Mapping for High Dynamic Range Images Using OpenCL with GPU},
author={Liao, Kuo-Feng and Hsu, Yarsun},
year={2017}
}
Tone mapping has been used to transfer HDR (high dynamic range) images to low dynamic range. This paper describes an algorithm to display high dynamic range images. Although local tone-mapping operator is better than global operator in reproducing images with better details and contrast, however, local tone mapping algorithm usually requires a huge amount of computation and it takes a long time to display an HDR image. We have designed a highly parallel method using Graphics Processing Unit (GPU) to accelerate the computation in order to achieve a real-time display. The algorithm can be highly parallelized. In order to run on different heterogeneous systems, we choose OpenCL, instead of CUDA, for our implementation. We have demonstrated the speed-up can be as high as 63 times for a 1280×960 image.
August 26, 2017 by hgpu
