- April, 2016: Paper and supplementary materials available (see Downloads).
- April, 2016: Web launched.
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.
- Paper [PDF, 27.1 MB]
- Supplementary material [PDF, 23.0 MB]
- Supplementary video [AVI, 22.4 MB]
- Slides [PPTX, 83.4 MB]
- Code [RAR, 103.0 MB]
The code provided is property of Universidad de Zaragoza - free for non-commercial purposes.
- 2016: Convolutional Sparse Coding for Capturing High Speed Video Content
- 2015: Compressive High-Speed Video Acquisition: Video Database
The authors would like to thank Karol Myszkowski, as well as Jose Echevarria and Adrian Jarabo, for fruitful insights and discussion. We would also like to thank Saghi Hajisharif and Jonas Unger, for sharing their results and for their assistance with them; Nicolas Landa for preliminary testing of traditional compressive sensing on HDR; and Maria Angeles Losada and the Photonic Technologies Group at Universidad de Zaragoza for their optical instrumentation. Ana Serrano was supported by an FPI grant from the Spanish Ministry of Economy and Competitivity (project Lightslice). Felix Heide was supported by a Four-year Fellowship from the University of British Columbia. Diego Gutierrez would like to acknowledge support from the BBVA Foundation and project Lightslice. Gordon Wetzstein was supported by a Terman Faculty Fellowship and by the Intel Strategic Research Alliance on Compressive Sensing. Belen Masia was partially supported by the Max Planck Center on Visual Computing and Communication.