New Paper: “Accelerating Translational Image Registration for HDR Images on GPU”

Hello academy-lovers!

Our new study is now available on arXiv:

K.C. Alpay, K.B. Aydemir, A. Temizel, Accelerating Translational Image Registration for HDR Images on GPU, arXiv:2007.06483, July 2020.

Our paper got accepted for publication in High Performance Computing Conference: BAŞARIM 2020.

It also got invited for submission to Concurrency and Computation: Practice and Experience | Wiley (CCPE) Journal BASARIM2020 Special Issue.

Click here to watch our conference presentation.

paper image
LDR alignment prevents blurry artifacts in the final HDR image.

Abstract: High Dynamic Range (HDR) images are generated using multiple exposures of a scene. When a hand-held camera is used to capture a static scene, these images need to be aligned by globally shifting each image in both dimensions. For a fast and robust alignment, the shift amount is commonly calculated using Median Threshold Bitmaps (MTB) and creating an image pyramid. In this study, we optimize these computations using a parallel processing approach utilizing GPU. Experimental evaluation shows that the proposed implementation achieves a speed-up of up to 6.24 times over the baseline multi-threaded CPU implementation on the alignment of one image pair. The source code is available at

Hope to see you in the next studies!

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.