多模态信息融合 - arXivDaily 专题

1609.08535 2016-09-28 cs.HC 71%

Chronodes: Interactive Multi-focus Exploration of Event Sequences

Peter J Polack, Shang-Tse Chen, Minsuk Kahng, Kaya de Barbaro, Moushumi Sharmin, Rahul Basole, Duen Horng Chau

专题命中多聚焦/多曝光融合：multi-focus(title)

2304.12748 2023-04-26 cs.CV 70%

Inverting the Imaging Process by Learning an Implicit Camera Model

Xin Huang, Qi Zhang, Ying Feng, Hongdong Li, Qing Wang

专题命中多聚焦/多曝光融合：multi-focus(abstract)；multi-exposure(abstract)；分类 cs.CV

Comments Accepted to CVPR 2023. Project page: https://xhuangcv.github.io/neucam/

2304.04464 2023-04-11 cs.CV 70%

Exposure Fusion for Hand-held Camera Inputs with Optical Flow and PatchMatch

Ru Li, Guanghui Liu, Bing Zeng, Shuaicheng Liu

专题命中多聚焦/多曝光融合：image fusion(abstract)；multi-exposure(abstract)；分类 cs.CV

2309.03827 2023-09-08 cs.CV cs.GR cs.LG cs.MM eess.IV 67%

ArtHDR-Net: Perceptually Realistic and Accurate HDR Content Creation

Hrishav Bakul Barua, Ganesh Krishnasamy, KokSheik Wong, Kalin Stefanov, Abhinav Dhall

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV、eess.IV、cs.MM

Comments Accepted in Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), Taipei, Taiwan

2510.21815 2025-10-28 eess.IV cs.AI cs.CV 62%

HDR Image Reconstruction using an Unsupervised Fusion Model

Kumbha Nagaswetha

机构 * Department of Electrical Communication Engineering, Indian Institute of Science（电子通信工程系，印度科学研究院）

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV、eess.IV

2506.23537 2025-07-08 eess.IV cs.CV 62%

AFUNet: Cross-Iterative Alignment-Fusion Synergy for HDR Reconstruction via Deep Unfolding Paradigm

Xinyue Li, Zhangkai Ni, Wenhan Yang

机构 * Tongji University（同济大学）； Pengcheng Laboratory（鹏城实验室）

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV、eess.IV

Comments Accepted to International Conference on Computer Vision (ICCV) 2025

2502.20824 2025-05-02 cs.CV eess.IV 62%

MFSR-GAN: Multi-Frame Super-Resolution with Handheld Motion Modeling

Fadeel Sher Khan, Joshua Ebenezer, Hamid Sheikh, Seok-Jun Lee

机构 * The University of Texas at Austin（德克萨斯大学奥斯汀分校）； Samsung Research America（三星美国研究实验室）

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV、eess.IV

Comments Accepted to NTIRE Workshop at CVPR 2025; 8 pages, 6 figures

2407.16308 2024-07-24 cs.CV eess.IV 62%

SAFNet: Selective Alignment Fusion Network for Efficient HDR Imaging

Lingtong Kong, Bo Li, Yike Xiong, Hao Zhang, Hong Gu, Jinwei Chen

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV、eess.IV

Comments Accepted by ECCV 2024

2201.02625 2024-07-24 eess.IV cs.CV 62%

FlexHDR: Modelling Alignment and Exposure Uncertainties for Flexible HDR Imaging

Sibi Catley-Chandar, Thomas Tanay, Lucas Vandroux, Aleš Leonardis, Gregory Slabaugh, Eduardo Pérez-Pellitero

专题命中多聚焦/多曝光融合：image fusion(abstract)；分类 cs.CV、eess.IV

Comments Accepted to IEEE Transactions on Image Processing (TIP) 2022

Journal ref IEEE Transactions on Image Processing, 2022, vol. 31, pp. 5923-5935

2310.16139 2024-04-26 eess.IV cs.CV 62%

Pix2HDR -- A pixel-wise acquisition and deep learning-based synthesis approach for high-speed HDR videos

Caixin Wang, Jie Zhang, Matthew A. Wilson, Ralph Etienne-Cummings

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV、eess.IV

Comments 17 pages, 18 figures

2203.03624 2023-09-08 eess.IV cs.CV 62%

FCNet: A Convolutional Neural Network for Arbitrary-Length Exposure Estimation

Jin Liang, Yuchen Yang, Anran Zhang, Jun Xu, Hui Li, Xiantong Zhen

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV、eess.IV

2208.02448 2023-05-30 cs.CV eess.IV 62%

Deep Progressive Feature Aggregation Network for High Dynamic Range Imaging

Jun Xiao, Qian Ye, Tianshan Liu, Cong Zhang, Kin-Man Lam

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV、eess.IV

2303.17253 2023-04-18 eess.IV cs.CV 62%

HDR Imaging with Spatially Varying Signal-to-Noise Ratios

Yiheng Chi, Xingguang Zhang, Stanley H. Chan

专题命中多聚焦/多曝光融合：image fusion(abstract)；分类 cs.CV、eess.IV

2212.03002 2022-12-07 cs.CV eess.IV 62%

Supervised Image Segmentation for High Dynamic Range Imaging

Ali Reza Omrani, Davide Moroni

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV、eess.IV

2211.12352 2022-11-24 cs.CV eess.IV 62%

GlowGAN: Unsupervised Learning of HDR Images from LDR Images in the Wild

Chao Wang, Ana Serrano, Xingang Pan, Bin Chen, Hans-Peter Seidel, Christian Theobalt, Karol Myszkowski, Thomas Leimkuehler

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV、eess.IV

2107.02299 2022-09-26 cs.CV eess.IV 62%

LightFuse: Lightweight CNN based Dual-exposure Fusion

Ziyi Liu, Jie Yang, Svetlana Yanushkevich, Orly Yadid-Pecht

专题命中多聚焦/多曝光融合：image fusion(abstract)；分类 cs.CV、eess.IV

2203.14341 2022-09-02 eess.IV cs.CV cs.LG 62%

MFSNet: A Multi Focus Segmentation Network for Skin Lesion Segmentation

Hritam Basak, Rohit Kundu, Ram Sarkar

专题命中多聚焦/多曝光融合：multi-focus(abstract)；分类 cs.CV、eess.IV

2102.01850 2022-07-18 eess.IV cs.CV 62%

UPHDR-GAN: Generative Adversarial Network for High Dynamic Range Imaging with Unpaired Data

Ru Li, Chuan Wang, Jue Wang, Guanghui Liu, Heng-Yu Zhang, Bing Zeng, Shuaicheng Liu

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV、eess.IV

Comments Accepted by IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)

Journal ref IEEE Transactions on Circuits and Systems for Video Technology, 2022

2111.05226 2022-05-06 eess.IV cs.CV 62%

Leveraging blur information for plenoptic camera calibration

Mathieu Labussière, Céline Teulière, Frédéric Bernardin, Omar Ait-Aider

专题命中多聚焦/多曝光融合：multi-focus(abstract)；分类 cs.CV、eess.IV

Comments arXiv admin note: text overlap with arXiv:2004.07745

2107.06211 2022-04-13 eess.IV cs.CV 62%

Attention-Guided Progressive Neural Texture Fusion for High Dynamic Range Image Restoration

Jie Chen, Zaifeng Yang, Tsz Nam Chan, Hui Li, Junhui Hou, Lap-Pui Chau

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV、eess.IV

2201.07066 2022-01-19 eess.IV cs.CV 62%

Joint denoising and HDR for RAW video sequences

A. Buades, O. Martorell, M. Sánchez-Beeckman

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV、eess.IV

Comments arXiv admin note: text overlap with arXiv:1812.11207

2110.04491 2021-10-12 eess.IV cs.CV 62%

Invertible Tone Mapping with Selectable Styles

Zhuming Zhang, Menghan Xia, Xueting Liu, Chengze Li, Tien-Tsin Wong

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV、eess.IV

1908.00186 2019-08-02 eess.IV cs.CV 62%

Single-Shot High Dynamic Range Imaging with Spatially Varying Exposures Considering Hue Distortion

Chihiro Go, Yuma Kinoshita, Sayaka Shiota, Hitoshi Kiya

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV、eess.IV

Comments To appear in 2019 IEEE 8th Global Conference on Consumer Electronics

2408.06083 2026-07-31 cs.CV 版本更新 57%

Towards Robust Monocular Depth Estimation in Non-Lambertian Surfaces

面向非朗伯表面的鲁棒单目深度估计

Junrui Zhang, Jiaqi Li, Yachuan Huang, Yiran Wang, Jinghong Zheng, Liao Shen, Zhiguo Cao

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV

AI总结该研究针对单目深度估计在非朗伯表面预测鲁棒性不足的问题，提出梯度域约束的区域引导方法及可选光照融合模块，在相关数据集和竞赛测试中取得优于Depth Anything V2的性能。

Comments Accepted to ECCV 2024 Workshop TRICKY

2607.17611 2026-07-21 cs.CV cs.AI 新提交 57%

Coarse-to-fine Framework for Generative MEF via Implicit Neural Representation

基于隐式神经表示的生成式多曝光融合粗到细框架

Sangmin Han, Jinho Kim, Jinwoo Kim, Dongyoung Kim, Seon Joo Kim

机构 * Yonsei University（延世大学）； LG Electronics（LG电子）

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV

AI总结该研究针对多曝光融合问题，提出LIIFusion粗到细框架，粗阶段通过自适应曝光校正进行低分辨率生成融合，细阶段将局部隐式图像函数适配为多曝光融合函数，加速达3.5倍，提升了结构保真度和感知质量，为生成式MEF实际应用提供途径。

2606.29845 2026-07-10 cs.CV 版本更新 57%

Bricker to BRACE: A Bracket Exposure RAW Dataset and Restoration Model for Flicker-Banding

Bricker到BRACE：用于闪烁条纹的括号曝光RAW数据集和恢复模型

Zihan Zhou, Libo Zhu, Jue Gong, Zhiyi Zhou, Jiezhang Cao, Yong Guo, Yulun Zhang

机构 * Shanghai Jiao Tong University（上海交通大学）； Huawei Consumer Business Group（华为消费者业务集团）

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV

AI总结针对屏幕拍摄图像中的闪烁条纹问题，提出基于多帧括号RAW的恢复模型BRACE，利用频率感知条纹先验和多尺度空间交叉注意力调制器，在合成和真实基准上达到最优性能。

2512.16357 2026-07-07 cs.CV 版本更新 57%

GMODiff: One-Step Gain Map Refinement with Diffusion Priors for HDR Reconstruction

GMODiff：基于扩散先验的单步增益图优化用于高动态范围重建

Tao Hu, Weiyu Zhou, Yanjie Tu, Peng Wu, Wei Dong, Qingsen Yan, Yanning Zhang

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV

AI总结本文提出GMODiff，通过单步扩散框架优化增益图，提升多曝光HDR重建的动态范围和效率，实验表明其性能优于现有方法且速度提升100倍。

Comments This paper is accepted by ECCV2026

2512.04390 2026-07-01 cs.CV cs.AI 版本更新 57%

FMA-Net++: Motion- and Exposure-Aware Joint Video Super-Resolution and Deblurring

FMA-Net++：运动与曝光感知的联合视频超分辨率与去模糊

Geunhyuk Youk, Jihyong Oh, Munchurl Kim

机构 * KAIST（韩国科学技术院）； CMLab, Chung-Ang University（中央大学CMLab）

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV

AI总结提出非循环序列级框架FMA-Net++，通过层次细化双向聚合块实现并行处理与长时域建模，引入曝光时间感知调制层处理曝光依赖模糊，在合成数据上训练达到SOTA性能。

Comments Accepted to ECCV 2026. Project Page: https://kaist-viclab.github.io/fmanetpp_site/

2605.14703 2026-05-15 cs.CV 57%

Generating HDR Video from SDR Video

从SDR视频生成HDR视频

SaiKiran Tedla, Francesco Banterle, Trevor Canham, Karanpreet Raja, David B. Lindell, Kiriakos N. Kutulakos, Jiacheng Li, Feiran Li, Daisuke Iso

机构 * Sony Research（索尼研究实验室）； York University（约克大学）； Vector Institute（向量研究所）； University of Toronto（多伦多大学）

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV

AI总结本文提出利用大规模生成视频模型从普通SDR视频生成HDR视频，引入多曝光视频模型和可学习视频合并模型，通过实验验证了其在真实场景和电影中的HDR转换能力。

2605.11115 2026-05-13 cs.CV cs.GR cs.LG 57%

LatentHDR: Decoupling Exposure from Diffusion via Conditional Latent-to-Latent Mapping for Text/Image-to-Panoramic HDR

LatentHDR: 通过条件潜变量到潜变量映射解耦曝光以生成文本/图像到全景HDR

Pedram Fekri, WenChen Li, William Chen, Peter Altamirano

机构 * Monks AI Research Lab（Monks AI研究院）

专题命中多聚焦/多曝光融合：multi-exposure(abstract)；分类 cs.CV

AI总结本文提出LatentHDR框架，通过在潜在空间中解耦场景生成与曝光建模，实现高效且结构一致的HDR生成，实验表明其在动态范围和感知质量上优于现有方法，并显著降低计算成本。

1. 多聚焦/多曝光融合 135 篇

Chronodes: Interactive Multi-focus Exploration of Event Sequences

Inverting the Imaging Process by Learning an Implicit Camera Model

Exposure Fusion for Hand-held Camera Inputs with Optical Flow and PatchMatch

ArtHDR-Net: Perceptually Realistic and Accurate HDR Content Creation

HDR Image Reconstruction using an Unsupervised Fusion Model

AFUNet: Cross-Iterative Alignment-Fusion Synergy for HDR Reconstruction via Deep Unfolding Paradigm

MFSR-GAN: Multi-Frame Super-Resolution with Handheld Motion Modeling

SAFNet: Selective Alignment Fusion Network for Efficient HDR Imaging

FlexHDR: Modelling Alignment and Exposure Uncertainties for Flexible HDR Imaging

Pix2HDR -- A pixel-wise acquisition and deep learning-based synthesis approach for high-speed HDR videos

FCNet: A Convolutional Neural Network for Arbitrary-Length Exposure Estimation

Deep Progressive Feature Aggregation Network for High Dynamic Range Imaging

HDR Imaging with Spatially Varying Signal-to-Noise Ratios

Supervised Image Segmentation for High Dynamic Range Imaging

GlowGAN: Unsupervised Learning of HDR Images from LDR Images in the Wild

LightFuse: Lightweight CNN based Dual-exposure Fusion

MFSNet: A Multi Focus Segmentation Network for Skin Lesion Segmentation

UPHDR-GAN: Generative Adversarial Network for High Dynamic Range Imaging with Unpaired Data

Leveraging blur information for plenoptic camera calibration

Attention-Guided Progressive Neural Texture Fusion for High Dynamic Range Image Restoration

Joint denoising and HDR for RAW video sequences

Invertible Tone Mapping with Selectable Styles

Single-Shot High Dynamic Range Imaging with Spatially Varying Exposures Considering Hue Distortion

Towards Robust Monocular Depth Estimation in Non-Lambertian Surfaces

Coarse-to-fine Framework for Generative MEF via Implicit Neural Representation

Bricker to BRACE: A Bracket Exposure RAW Dataset and Restoration Model for Flicker-Banding

GMODiff: One-Step Gain Map Refinement with Diffusion Priors for HDR Reconstruction

FMA-Net++: Motion- and Exposure-Aware Joint Video Super-Resolution and Deblurring

Generating HDR Video from SDR Video

LatentHDR: Decoupling Exposure from Diffusion via Conditional Latent-to-Latent Mapping for Text/Image-to-Panoramic HDR