Convolutional Sparse Coding for High Dynamic Range Imaging. (27th May 2016)

• Record Type: Journal Article
• Authors: Serrano, Ana; Heide, Felix; Gutierrez, Diego; Wetzstein, Gordon; Masia, Belen
• Is Part Of: Computer graphics forum. Volume 35:Number 2(2016)
• Page Start: 153
• Page End: 163
• Publication Date: 2016-05-27
• Subjects: Categories and Subject Descriptors (according to ACM CCS) -- I.4.1 [Image Processing and Computer Vision] -- Digitization and Image Capture—; Computer graphics -- Periodicals; 006.605
• DOI: 10.1111/cgf.12819
• Languages: English
• ISSNs: 0167-7055

Abstract

Abstract: Current HDR acquisition techniques are based on either (i) fusing multibracketed, low dynamic range (LDR) images, (ii) modifying existing hardware and capturing different exposures simultaneously with multiple sensors, or (iii) reconstructing a single image with spatially‐varying pixel exposures. In this paper, we propose a novel algorithm to recover high‐quality HDRI images from a single, coded exposure. The proposed reconstruction method builds on recently‐introduced ideas of convolutional sparse coding (CSC); this paper demonstrates how to make CSC practical for HDR imaging. We demonstrate that the proposed algorithm achieves higher‐quality reconstructions than alternative methods, we evaluate optical coding schemes, analyze algorithmic parameters, and build a prototype coded HDR camera that demonstrates the utility of convolutional sparse HDRI coding with a custom hardware platform.