Unsupervised Single-Image High Dynamic Range Rendering via Multi-Exposure Priors
| |||||||||||||
Han Wang1, Bolun Zheng1, Quan Chen2, Qianyu Zhang1, Tao Zhang1, Jiyong Zhang1, Xiang Tian3. Unsupervised Single-Image High Dynamic Range Rendering via Multi-Exposure Priors[J]. Journal of Zhejiang University Science C, 1998, -1(-1): .
@article{title="Unsupervised Single-Image High Dynamic Range Rendering via Multi-Exposure Priors",
author="Han Wang1, Bolun Zheng1, Quan Chen2, Qianyu Zhang1, Tao Zhang1, Jiyong Zhang1, Xiang Tian3",
journal="Journal of Zhejiang University Science C",
volume="-1",
number="-1",
pages="",
year="1998",
publisher="Zhejiang University Press & Springer",
doi="10.1631/ENG.ITEE.2025.0116"
}
%0 Journal Article
%T Unsupervised Single-Image High Dynamic Range Rendering via Multi-Exposure Priors
%A Han Wang1
%A Bolun Zheng1
%A Quan Chen2
%A Qianyu Zhang1
%A Tao Zhang1
%A Jiyong Zhang1
%A Xiang Tian3
%J Journal of Zhejiang University SCIENCE C
%V -1
%N -1
%P
%@ 1869-1951
%D 1998
%I Zhejiang University Press & Springer
%DOI 10.1631/ENG.ITEE.2025.0116
TY - JOUR
T1 - Unsupervised Single-Image High Dynamic Range Rendering via Multi-Exposure Priors
A1 - Han Wang1
A1 - Bolun Zheng1
A1 - Quan Chen2
A1 - Qianyu Zhang1
A1 - Tao Zhang1
A1 - Jiyong Zhang1
A1 - Xiang Tian3
J0 - Journal of Zhejiang University Science C
VL - -1
IS - -1
SP -
EP - 0
%@ 1869-1951
Y1 - 1998
PB - Zhejiang University Press & Springer
ER -
DOI - 10.1631/ENG.ITEE.2025.0116
Abstract: Reconstructing high dynamic range (HDR) images from a single low dynamic range (LDR) input requires recovering
missing information in highlight-clipped and shadow-distorted regions. Existing methods generally rely on sufficient groundtruth HDR images as supervision signals or multi-exposure LDR sequences to improve quality, limiting their flexibility. To
address this, we propose USME-HDR, a framework for single-image HDR reconstruction based on multi-exposure priors, where
the HDR reconstruction stage is learned without ground-truth HDR supervision. Specifically, an Exposure-Adjustment Network
(EAN) is trained in a supervised manner to map a single LDR image to over/under-exposure pairs. Inspired by Retinex theory,
we further decompose the input into Light Map and Light Feature, which are fed into EAN as auxiliary inputs for luminanceaware exposure generation. An exposure ratio guidance mechanism is further introduced to improve luminance fidelity. Finally,
the HDR image is synthesized by fusing the original LDR image with generated multi-exposure images, refined through selfsupervised optimization. Experiments demonstrate that, during the testing phase, USME-HDR reconstructs visually compelling
HDR images from only a single LDR input, without requiring real low- or high-exposure images.
Full Text:
<3>
CLC number:
On-line Access: 2026-05-07
Received: 2025-11-04
Revision Accepted: 2026-04-19
Crosschecked: 0000-00-00
Cited: 0
Clicked: 10
Open peer comments: Debate/Discuss/Question/Opinion
<1>