High Dynamic Range Video Synthesis Using Superpixel-Based Illuminance-Invariant Motion Estimation

Abstract

We propose a robust high dynamic range (HDR) video synthesis algorithm using the superpixel-based illuminance-invariant motion estimation technique. The proposed algorithm first selects an input frame in an alternating exposed input video as the reference. Then, the correspondences between two adjacent frames are estimated by employing a feature descriptor, which is robust against illuminance variation, and a superpixel segmentation technique. Next, the input frames are warped to the reference frame using the estimated motion maps. Finally, the final HDR frame is synthesized by constructing a weight map, which can handle complex motions and poor exposures by considering the underlying structures in the input frames. Experimental results on real test sequences show that the proposed algorithm can provide high-quality HDR videos compared with those obtained by state-of-the-art algorithms in terms of both subjective and objective evaluations. INDEX TERMS High dynamic range (HDR) video synthesis, HDR imaging, motion estimation.

A Compact Multi-Exposure File Format for Backward and Forward Compatible HDR Imaging

“…Secondly, the process of capturing bracketed exposures may result in misalignment artifacts due to camera and/or object motion. Although these artifacts could be mitigated by the application of a deghosting algorithm [14,15,16], complete elimination of ghosting artifacts remains to be an elusive problem. As such, photographers may choose to preserve the input exposures and use them to create an HDR image with better quality as new and improved deghosting algorithms emerge.…”

“…The difference-based CES, patch-based CES, and the ResCES outcomes are evaluated across the fold-1 test set(15,26,40,43, and 50) using the SSIM and PSNR metrics. The proposed ResCES algorithm shows noticeable improvement in all cases.…”