Single-Image HDR Reconstruction by Learning to Reverse the Camera Pipeline
Authors:
Top Cited Papers
- 1 June 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1648-1657
Abstract
Recovering a high dynamic range (HDR) image from a single low dynamic range (LDR) input image is challenging due to missing details in under-/over-exposed regions caused by quantization and saturation of camera sensors. In contrast to existing learning-based methods, our core idea is to incorporate the domain knowledge of the LDR image formation pipeline into our model. We model the HDR-to-LDR image formation pipeline as the (1) dynamic range clipping, (2) non-linear mapping from a camera response function, and (3) quantization. We then propose to learn three specialized CNNs to reverse these steps. By decomposing the problem into specific sub-tasks, we impose effective physical constraints to facilitate the training of individual sub-networks. Finally, we jointly fine-tune the entire model end-to-end to reduce error accumulation. With extensive quantitative and qualitative experiments on diverse image datasets, we demonstrate that the proposed method performs favorably against state-of-the-art single-image HDR reconstruction algorithms.Keywords
All Related Versions
This publication has 33 references indexed in Scilit:
- RAISE2015
- Modeling Radiometric Uncertainty for Vision with Tone-Mapped Color ImagesIEEE Transactions on Pattern Analysis and Machine Intelligence, 2014
- Physiological inverse tone mapping based on retina responseThe Visual Computer, 2013
- Ghost detection and removal for high dynamic range images: Recent advancesSignal Processing: Image Communication, 2012
- High dynamic range video with ghost removal2010
- Estimating Radiometric Response Functions from Image Noise Variance2008
- Radiometric Calibration from Noise Distributions2007
- A framework for inverse tone mappingThe Visual Computer, 2007
- Radiometric calibration from a single image2004
- Image quilting for texture synthesis and transfer2001