pith. machine review for the scientific record. sign in

arxiv: 2604.14632 · v1 · submitted 2026-04-16 · 💻 cs.CV

Recognition: unknown

High-Speed Full-Color HDR Imaging via Unwrapping Modulo-Encoded Spike Streams

Chu Zhou , Siqi Yang , Kailong Zhang , Heng Guo , Zhaofei Yu , Boxin Shi , Imari Sato
Authors on Pith no claims yet

Pith reviewed 2026-05-10 12:22 UTC · model grok-4.3

classification 💻 cs.CV
keywords HDR imagingModulo sensorsSpike camerasHigh-speed imagingDynamic rangeUnwrapping algorithmGenerative priorsComputational photography
0
0 comments X

The pith

An exposure-decoupled modulo formulation and iteration-free unwrapping enable 1000 FPS full-color HDR imaging from spike streams.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

Conventional HDR methods trade motion artifacts from multiple exposures against information loss in single captures. Modulo sensors wrap high dynamic range into bounded measurements but have been limited by slow iterative unwrapping and grayscale hardware. The paper establishes an exposure-decoupled formulation that interleaves multiple measurements in time while keeping a clean per-observation model, then pairs it with a non-iterative unwrapping step that folds diffusion generative priors together with the physical least absolute remainder property. This combination supports artifact-free HDR reconstruction at high speed. A reader should care because it removes the speed and color barriers that have kept modulo imaging out of practical dynamic capture.

Core claim

The central claim is that an exposure-decoupled formulation of modulo imaging permits temporal interleaving of multiple measurements under a clean observation model, and that an iteration-free unwrapping algorithm integrating diffusion-based generative priors with the least absolute remainder property produces physics-consistent HDR images from these measurements. The authors validate the approach through a proof-of-concept hardware system that records modulo-encoded spike streams, achieving 1000 FPS full-color HDR while lowering output bandwidth from roughly 20 Gbps to 6 Gbps.

What carries the argument

The exposure-decoupled modulo imaging formulation together with the iteration-free unwrapping algorithm that combines diffusion generative priors and the least absolute remainder property.

If this is right

  • Full-color HDR capture at 1000 frames per second becomes practical for dynamic scenes.
  • Data bandwidth falls from approximately 20 Gbps to 6 Gbps while preserving native spike-camera temporal resolution.
  • Modulo imaging can move from low-speed grayscale demonstrations to real-time color use cases.
  • The hardware proof-of-concept shows that spike-stream modulo encoding is deployable without previous speed or color limits.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The interleaving idea may transfer to other high-speed sensors that currently suffer from exposure trade-offs.
  • Pairing generative priors with explicit physical constraints could simplify reconstruction in related computational imaging tasks such as phase unwrapping or event-based vision.
  • Bandwidth savings at high frame rates suggest the pipeline could scale to higher spatial resolutions or multi-camera arrays without proportional storage growth.

Load-bearing premise

The iteration-free unwrapping algorithm that merges diffusion-based generative priors with the physical least absolute remainder property will produce artifact-free and physics-consistent HDR images from the interleaved modulo measurements.

What would settle it

Direct comparison of the reconstructed HDR frames against known ground-truth high-dynamic-range video in fast-moving high-contrast scenes; visible artifacts, temporal inconsistencies, or deviation from physical light levels would disprove the claim.

Figures

Figures reproduced from arXiv: 2604.14632 by Boxin Shi, Chu Zhou, Heng Guo, Imari Sato, Kailong Zhang, Siqi Yang, Zhaofei Yu.

Figure 1
Figure 1. Figure 1: We present a complete modulo-based HDR imaging system capable of high-speed, full-color HDR acquisition. (a) Algorithm comparison on the [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Illustration of the difference between modulo imaging formulations. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Illustration of the proposed iteration-free modulo unwrapping framework, which consists of two stages: (a) diffusion-based HDR prior extraction [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Illustration of the least absolute remainder (LAR) property of modulo [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Illustration of the proposed bandwidth-efficient hardware implementation, designed as a proof-of-concept to validate the practical viability of our [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Qualitative comparisons on synthetic data using the UnModNet dataset [60] between our algorithm and state-of-the-art HDR reconstruction approaches, [PITH_FULL_IMAGE:figures/full_fig_p011_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Comparisons on pedestrian detection using reconstructed HDR driving scenes between our algorithm and state-of-the-art HDR reconstruction approaches, [PITH_FULL_IMAGE:figures/full_fig_p012_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Qualitative comparisons on real-world dynamic scenes captured by our proof-of-concept hardware prototype. We compare our algorithm with two [PITH_FULL_IMAGE:figures/full_fig_p013_8.png] view at source ↗
read the original abstract

Conventional RGB-based high dynamic range (HDR) imaging faces a fundamental trade-off between motion artifacts in multi-exposure captures and irreversible information loss in single-shot techniques. Modulo sensors offer a promising alternative by encoding theoretically unbounded dynamic range into wrapped measurements. However, existing modulo solutions remain bottlenecked by iterative unwrapping overhead and hardware constraints limiting them to low-speed, grayscale capture. In this work, we present a complete modulo-based HDR imaging system that enables high-speed, full-color HDR acquisition by synergistically advancing both the sensing formulation and the unwrapping algorithm. At the core of our approach is an exposure-decoupled formulation of modulo imaging that allows multiple measurements to be interleaved in time, preserving a clean, observation-wise measurement model. Building upon this, we introduce an iteration-free unwrapping algorithm that integrates diffusion-based generative priors with the physical least absolute remainder property of modulo images, supporting highly efficient, physics-consistent HDR reconstruction. Finally, to validate the practical viability of our system, we demonstrate a proof-of-concept hardware implementation based on modulo-encoded spike streams. This setup preserves the native high temporal resolution of spike cameras, achieving 1000 FPS full-color imaging while reducing output data bandwidth from approximately 20 Gbps to 6 Gbps. Extensive evaluations indicate that our coordinated approach successfully overcomes key systemic bottlenecks, demonstrating the feasibility of deploying modulo imaging in dynamic scenarios.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript claims to present a complete modulo-based HDR imaging system enabling high-speed full-color acquisition. It introduces an exposure-decoupled formulation of modulo imaging to support interleaved multi-measurement capture while preserving per-observation models, an iteration-free unwrapping algorithm that fuses diffusion-based generative priors with the physical least-absolute-remainder property for artifact-free HDR reconstruction, and a proof-of-concept spike-camera hardware prototype that achieves 1000 FPS full-color imaging while reducing output bandwidth from ~20 Gbps to 6 Gbps.

Significance. If the reported performance and reconstruction quality hold under rigorous validation, the work would represent a meaningful advance in computational imaging by removing the motion-artifact versus dynamic-range trade-off that limits conventional RGB HDR methods and by demonstrating practical deployment of modulo sensors in dynamic, high-speed scenarios. The hardware prototype and explicit bandwidth-reduction numbers constitute concrete, falsifiable contributions that could influence sensor design and real-time vision pipelines.

major comments (2)
  1. [Abstract] The central feasibility claim (1000 FPS full-color HDR with 6 Gbps output) rests on the iteration-free unwrapping algorithm, yet the abstract supplies no derivation, pseudocode, or quantitative metrics (PSNR, SSIM, error bars, ablation on diffusion prior strength) for how the diffusion model is conditioned on the least-absolute-remainder property or how consistency with the physical measurement model is enforced. Without these details the physics-consistency guarantee cannot be assessed.
  2. [Abstract] The exposure-decoupled formulation is presented as enabling clean interleaving of measurements, but no explicit observation model, noise model, or proof that the modulo wrapping remains independent across interleaved exposures is supplied; this is load-bearing for the claim that the approach overcomes iterative unwrapping overhead.
minor comments (1)
  1. [Abstract] The abstract states 'extensive evaluations indicate success' without referencing any table, figure, or section containing the supporting data; this should be cross-referenced to the results section for reader convenience.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback on our manuscript. We address each major comment below and have revised the abstract to improve clarity while preserving its concise nature.

read point-by-point responses

  1. Referee: [Abstract] The central feasibility claim (1000 FPS full-color HDR with 6 Gbps output) rests on the iteration-free unwrapping algorithm, yet the abstract supplies no derivation, pseudocode, or quantitative metrics (PSNR, SSIM, error bars, ablation on diffusion prior strength) for how the diffusion model is conditioned on the least-absolute-remainder property or how consistency with the physical measurement model is enforced. Without these details the physics-consistency guarantee cannot be assessed.

    Authors: The abstract is a high-level summary. The full derivation of the iteration-free unwrapping algorithm, its conditioning on the least-absolute-remainder property, enforcement of physical consistency via the diffusion prior, pseudocode, and quantitative metrics including PSNR, SSIM, error bars, and ablations are provided in the main manuscript. We have revised the abstract to briefly note these components and the resulting physics-consistent reconstruction. revision: yes

  2. Referee: [Abstract] The exposure-decoupled formulation is presented as enabling clean interleaving of measurements, but no explicit observation model, noise model, or proof that the modulo wrapping remains independent across interleaved exposures is supplied; this is load-bearing for the claim that the approach overcomes iterative unwrapping overhead.

    Authors: The abstract summarizes the contribution at a high level. The explicit observation model, noise model, and proof of independence of modulo wrapping across interleaved exposures are formally derived in the manuscript. This decoupled formulation directly enables the non-iterative unwrapping. We have revised the abstract to include a concise reference to the observation model. revision: yes

Circularity Check

0 steps flagged

No significant circularity detected in derivation chain

full rationale

The paper grounds its core contributions—an exposure-decoupled modulo formulation, an iteration-free unwrapping method that combines external diffusion-based generative priors with the independently stated least-absolute-remainder property, and a spike-camera hardware prototype—directly in physical measurement models and external priors rather than self-referential definitions, fitted parameters renamed as predictions, or load-bearing self-citations. No equations or claims in the provided abstract or described components reduce the reported 1000 FPS / 6 Gbps feasibility result to the inputs by construction; the approach remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review yields no identifiable free parameters, axioms, or invented entities; the central claim rests on an unelaborated physical remainder property and external diffusion models whose integration details are not supplied.

pith-pipeline@v0.9.0 · 5563 in / 1159 out tokens · 47728 ms · 2026-05-10T12:22:35.105768+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

62 extracted references · 7 canonical work pages · 3 internal anchors

  1. [1]

    Single shot high dynamic range imaging using piecewise linear estimators

    Cecilia Aguerrebere, Andr ´es Almansa, Yann Gousseau, Julie Delon, and Pablo Muse. Single shot high dynamic range imaging using piecewise linear estimators. InProc. of International Conference on Computational Photography, pages 1–10, 2014

    2014

  2. [2]

    Reconfigurable snapshot HDR imaging using coded masks and inception network

    Masheal M Alghamdi, Qiang Fu, Ali Kassem Thabet, and Wolfgang Heidrich. Reconfigurable snapshot HDR imaging using coded masks and inception network. InVision, Modeling and Visualization, 2019

    2019

  3. [3]

    Deep plug-and-play algorithm for unsaturated imaging

    Jorge Bacca, Brayan Monroy, and Henry Arguello. Deep plug-and-play algorithm for unsaturated imaging. InProc. of International Conference on Acoustics, Speech and Signal Processing, pages 2460–2464, 2024. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 14

    2024

  4. [4]

    Inverse tone mapping

    Francesco Banterle, Patrick Ledda, Kurt Debattista, and Alan Chalmers. Inverse tone mapping. InProc. of International Conference on Computer Graphics and Interactive Techniques in Australasia and Southeast Asia, 2006

    2006

  5. [5]

    A 240 × 180 130 dB 3 µs latency global shutter spatiotemporal vision sensor.IEEE Journal of Solid-State Circuits, 49(10):2333–2341, 2014

    Christian Brandli, Raphael Berner, Minhao Yang, Shih-Chii Liu, and Tobi Delbruck. A 240 × 180 130 dB 3 µs latency global shutter spatiotemporal vision sensor.IEEE Journal of Solid-State Circuits, 49(10):2333–2341, 2014

    2014

  6. [6]

    1000 PFS HDR video with a spike-RGB hybrid camera

    Yakun Chang, Chu Zhou, Yuchen Hong, Liwen Hu, Chao Xu, Tiejun Huang, and Boxin Shi. 1000 PFS HDR video with a spike-RGB hybrid camera. InProc. of Computer Vision and Pattern Recognition, pages 22180–22190, 2023

    2023

  7. [7]

    Robust unfolding network for HDR imaging with modulo cameras

    Zhile Chen and Hui Ji. Robust unfolding network for HDR imaging with modulo cameras. InProc. of International Conference on Computer Vision, pages 25218–25228, 2025

    2025

  8. [8]

    UltraFusion: Ultra high dynamic imaging using exposure fusion

    Zixuan Chen, Yujin Wang, Xin Cai, Zhiyuan You, Zheming Lu, Fan Zhang, Shi Guo, and Tianfan Xue. UltraFusion: Ultra high dynamic imaging using exposure fusion. InProc. of Computer Vision and Pattern Recognition, pages 16111–16121, 2025

    2025

  9. [9]

    Recovering high dynamic range radiance maps from photographs

    Paul E Debevec and Jitendra Malik. Recovering high dynamic range radiance maps from photographs. InProc. of ACM SIGGRAPH, pages 369–378, 1997

    1997

  10. [10]

    Intrinsic single-image HDR reconstruction

    Sebastian Dille, Chris Careaga, and Ya ˘gız Aksoy. Intrinsic single-image HDR reconstruction. InProc. of European Conference on Computer Vision, pages 161–177, 2024

    2024

  11. [11]

    An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weis- senborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, et al. An image is worth 16 × 16 words: Transformers for image recognition at scale.arXiv preprint arXiv:2010.11929, 2020

    work page internal anchor Pith review Pith/arXiv arXiv 2010

  12. [12]

    Mantiuk, and Jonas Unger

    Gabriel Eilertsen, Joel Kronander, Gyorgy Denes, Rafał K. Mantiuk, and Jonas Unger. HDR image reconstruction from a single exposure using deep CNNs.ACM Transactions on Graphics (Proc. of ACM SIGGRAPH Asia), 36(6):178:1–178:15, 2017

    2017

  13. [13]

    Deep reverse tone mapping.ACM Transactions on Graphics (Proc

    Yuki Endo, Yoshihiro Kanamori, and Jun Mitani. Deep reverse tone mapping.ACM Transactions on Graphics (Proc. of ACM SIGGRAPH Asia), 36(6):177:1–177:10, 2017

    2017

  14. [14]

    Hybrid high dynamic range imaging fusing neuromorphic and conventional images

    Jin Han, Yixin Yang, Peiqi Duan, Chu Zhou, Lei Ma, Chao Xu, Tiejun Huang, Imari Sato, and Boxin Shi. Hybrid high dynamic range imaging fusing neuromorphic and conventional images. 2023

    2023

  15. [15]

    Neuromorphic camera guided high dynamic range imaging

    Jin Han, Chu Zhou, Peiqi Duan, Yehui Tang, Chang Xu, Chao Xu, Tiejun Huang, and Boxin Shi. Neuromorphic camera guided high dynamic range imaging. InProc. of Computer Vision and Pattern Recognition, pages 1730–1739, 2020

    2020

  16. [16]

    Comparison of single image HDR reconstruction methods- the caveats of quality assessment

    Param Hanji, Rafal Mantiuk, Gabriel Eilertsen, Saghi Hajisharif, and Jonas Unger. Comparison of single image HDR reconstruction methods- the caveats of quality assessment. InProc. of ACM SIGGRAPH, pages 1–8, 2022

    2022

  17. [17]

    Burst photography for high dynamic range and low-light imaging on mobile cameras.ACM Transactions on Graphics (Proc

    Samuel W Hasinoff, Dillon Sharlet, Ryan Geiss, Andrew Adams, Jonathan T Barron, Florian Kainz, Jiawen Chen, and Marc Levoy. Burst photography for high dynamic range and low-light imaging on mobile cameras.ACM Transactions on Graphics (Proc. of ACM SIGGRAPH), 35(6):1–12, 2016

    2016

  18. [18]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. InProc. of Computer Vision and Pattern Recognition, pages 770–778, 2016

    2016

  19. [19]

    Reducing the dimen- sionality of data with neural networks.Science, 313(5786):504–507, 2006

    Geoffrey E Hinton and Ruslan R Salakhutdinov. Reducing the dimen- sionality of data with neural networks.Science, 313(5786):504–507, 2006

    2006

  20. [20]

    Denoising diffusion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffusion probabilistic models. InProc. of Advances in Neural Information Processing Systems, pages 6840–6851, 2020

    2020

  21. [21]

    Generating content for HDR deghosting from frequency view

    Tao Hu, Qingsen Yan, Yuankai Qi, and Yanning Zhang. Generating content for HDR deghosting from frequency view. InProc. of Computer Vision and Pattern Recognition, pages 25732–25741, 2024

    2024

  22. [22]

    Analysis of the phase unwrapping algorithm.Applied Optics, 21(14):2470–2470, 1982

    Kazuyoshi Itoh. Analysis of the phase unwrapping algorithm.Applied Optics, 21(14):2470–2470, 1982

    1982

  23. [23]

    High dynamic range imaging using deep image priors

    Gauri Jagatap and Chinmay Hegde. High dynamic range imaging using deep image priors. InProc. of International Conference on Acoustics, Speech and Signal Processing, pages 9289–9293, 2020

    2020

  24. [24]

    Deep high dynamic range imaging of dynamic scenes.ACM Transactions on Graphics (Proc

    Nima Khademi Kalantari and Ravi Ramamoorthi. Deep high dynamic range imaging of dynamic scenes.ACM Transactions on Graphics (Proc. of ACM SIGGRAPH), 36(4):144–1, 2017

    2017

  25. [25]

    Deep HDR video from sequences with alternating exposures

    Nima Khademi Kalantari and Ravi Ramamoorthi. Deep HDR video from sequences with alternating exposures. InComputer Graphics F orum, volume 38, pages 193–205, 2019

    2019

  26. [26]

    Ghost removal in high dynamic range images

    Erum Arif Khan, Ahmet Oguz Akyuz, and Erik Reinhard. Ghost removal in high dynamic range images. InProc. of International Conference on Image Processing, pages 2005–2008, 2006

    2005

  27. [27]

    AFUNet: Cross-iterative alignment-fusion synergy for HDR reconstruction via deep unfolding paradigm

    Xinyue Li, Zhangkai Ni, and Wenhan Yang. AFUNet: Cross-iterative alignment-fusion synergy for HDR reconstruction via deep unfolding paradigm. InProc. of International Conference on Computer Vision, pages 10666–10675, 2025

    2025

  28. [28]

    Benchmarking detection transfer learning with vision Transformers.arXiv preprint arXiv:2111.11429, 2021

    Yanghao Li, Saining Xie, Xinlei Chen, Piotr Dollar, Kaiming He, and Ross Girshick. Benchmarking detection transfer learning with vision Transformers.arXiv preprint arXiv:2111.11429, 2021

    work page arXiv 2021

  29. [29]

    Lightweight video denoising using aggregated shifted window attention

    Lydia Lindner, Alexander Effland, Filip Ilic, Thomas Pock, and Erich Kobler. Lightweight video denoising using aggregated shifted window attention. InProc. of Winter Conference on Applications of Computer Vision, pages 351–360, 2023

    2023

  30. [30]

    Joint HDR denoising and fusion: A real-world mobile HDR image dataset

    Shuaizheng Liu, Xindong Zhang, Lingchen Sun, Zhetong Liang, Hui Zeng, and Lei Zhang. Joint HDR denoising and fusion: A real-world mobile HDR image dataset. InProc. of Computer Vision and Pattern Recognition, pages 13966–13975, 2023

    2023

  31. [31]

    Single-image HDR reconstruction by learning to reverse the camera pipeline

    Yu-Lun Liu, Wei-Sheng Lai, Yu-Sheng Chen, Yi-Lung Kao, Ming- Hsuan Yang, Yung-Yu Chuang, and Jia-Bin Huang. Single-image HDR reconstruction by learning to reverse the camera pipeline. InProc. of Computer Vision and Pattern Recognition, pages 1651–1660, 2020

    2020

  32. [32]

    SGDR: Stochastic Gradient Descent with Warm Restarts

    Ilya Loshchilov and Frank Hutter. SGDR: Stochastic gradient descent with warm restarts.arXiv preprint arXiv:1608.03983, 2016

    work page Pith review arXiv 2016

  33. [33]

    Decoupled Weight Decay Regularization

    Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization. arXiv preprint arXiv:1711.05101, 2017

    work page internal anchor Pith review Pith/arXiv arXiv 2017

  34. [34]

    2304.13625 , archivePrefix=

    Rafal K Mantiuk, Dounia Hammou, and Param Hanji. HDR-VDP-3: A multi-metric for predicting image differences, quality and contrast distortions in high dynamic range and regular content.arXiv preprint arXiv:2304.13625, 2023

    work page arXiv 2023

  35. [35]

    ExpandNet: A deep convolutional neural network for high dynamic range expansion from low dynamic range content

    Demetris Marnerides, Thomas Bashford-Rogers, Jonathan Hatchett, and Kurt Debattista. ExpandNet: A deep convolutional neural network for high dynamic range expansion from low dynamic range content. Computer Graphics F orum, 37:37–49, 2018

    2018

  36. [36]

    Fleming, Olga Sorkine, and Diego Gutierrez

    Belen Masia, Sandra Agustin, Roland W. Fleming, Olga Sorkine, and Diego Gutierrez. Evaluation of reverse tone mapping through varying exposure conditions.ACM Transactions on Graphics (Proc. of ACM SIGGRAPH Asia), 28(5):160:1–160:8, 2009

    2009

  37. [37]

    Deep optics for single-shot high-dynamic-range imaging

    Christopher A Metzler, Hayato Ikoma, Yifan Peng, and Gordon Wetzstein. Deep optics for single-shot high-dynamic-range imaging. InProc. of Computer Vision and Pattern Recognition, pages 1375–1385, 2020

    2020

  38. [38]

    High dynamic range imaging: Spatially varying pixel exposures

    Shree K Nayar and Tomoo Mitsunaga. High dynamic range imaging: Spatially varying pixel exposures. InProc. of Computer Vision and Pattern Recognition, pages 472–479, 2000

    2000

  39. [39]

    HDR-GAN: HDR image reconstruction from multi-exposed LDR images with large motions.IEEE Transactions on Image Processing, 30:3885–3896, 2021

    Yuzhen Niu, Jianbin Wu, Wenxi Liu, Wenzhong Guo, and Rynson WH Lau. HDR-GAN: HDR image reconstruction from multi-exposed LDR images with large motions.IEEE Transactions on Image Processing, 30:3885–3896, 2021

    2021

  40. [40]

    Robust high dynamic range imaging by rank minimization.IEEE Transactions on Pattern Analysis and Machine Intelligence, 37(6):1219–1232, 2014

    Tae-Hyun Oh, Joon-Young Lee, Yu-Wing Tai, and In So Kweon. Robust high dynamic range imaging by rank minimization.IEEE Transactions on Pattern Analysis and Machine Intelligence, 37(6):1219–1232, 2014

    2014

  41. [41]

    Towards practical and efficient high- resolution HDR deghosting with CNN

    K Ram Prabhakar, Susmit Agrawal, Durgesh Kumar Singh, Balraj Ashwath, and R Venkatesh Babu. Towards practical and efficient high- resolution HDR deghosting with CNN. InProc. of European Conference on Computer Vision, pages 497–513, 2020

    2020

  42. [42]

    High speed and high dynamic range video with an event camera.IEEE Transactions on Pattern Analysis and Machine Intelligence, 43(6):1964– 1980, 2019

    Henri Rebecq, Ren ´e Ranftl, Vladlen Koltun, and Davide Scaramuzza. High speed and high dynamic range video with an event camera.IEEE Transactions on Pattern Analysis and Machine Intelligence, 43(6):1964– 1980, 2019

    1964

  43. [43]

    Rempel, Matthew Trentacoste, Helge Seetzen, H

    Allan G. Rempel, Matthew Trentacoste, Helge Seetzen, H. David Young, Wolfgang Heidrich, Lorne Whitehead, and Greg Ward. LDR2HDR: On-the-fly reverse tone mapping of legacy video and photographs.ACM Transactions on Graphics (Proc. of ACM SIGGRAPH), 26(3), 2007

    2007

  44. [44]

    High-resolution image synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj ¨orn Ommer. High-resolution image synthesis with latent diffusion models. InProc. of Computer Vision and Pattern Recognition, pages 10684–10695, 2022

    2022

  45. [45]

    Single image HDR reconstruction using a CNN with masked features and perceptual loss.ACM Transactions on Graphics (Proc

    Marcel Santana Santos, Tsang Ing Ren, and Nima Khademi Kalantari. Single image HDR reconstruction using a CNN with masked features and perceptual loss.ACM Transactions on Graphics (Proc. of ACM SIGGRAPH), 39(4):80–1, 2020

    2020

  46. [46]

    High dynamic range video by spatially non-regular optical filtering

    Michael Sch ¨oberl, Alexander Belz, J ¨urgen Seiler, Siegfried Foessel, and Andr´e Kaup. High dynamic range video by spatially non-regular optical filtering. InProc. of International Conference on Image Processing, pages 2757–2760, 2012

    2012

  47. [47]

    Goldman, and Eli Shechtman

    Pradeep Sen, Nima Khademi Kalantari, Maziar Yaesoubi, Soheil Darabi, Dan B. Goldman, and Eli Shechtman. Robust patch-based HDR reconstruction of dynamic scenes.ACM Transactions on Graphics (Proc. of ACM SIGGRAPH), 31(6):203:1–203:11, 2012

    2012

  48. [48]

    Convolutional sparse coding for high dynamic range imaging

    Ana Serrano, Felix Heide, Diego Gutierrez, Gordon Wetzstein, and Belen Masia. Convolutional sparse coding for high dynamic range imaging. In Computer Graphics F orum, pages 153–163, 2016. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 15

    2016

  49. [49]

    Denoising Diffusion Implicit Models

    Jiaming Song, Chenlin Meng, and Stefano Ermon. Denoising diffusion implicit models.arXiv preprint arXiv:2010.02502, 2020

    work page internal anchor Pith review Pith/arXiv arXiv 2010

  50. [50]

    Instance Normalization: The Missing Ingredient for Fast Stylization

    Dmitry Ulyanov, Andrea Vedaldi, and Victor Lempitsky. Instance normalization: The missing ingredient for fast stylization.arXiv preprint arXiv:1607.08022, 2016

    work page Pith review arXiv 2016

  51. [51]

    LEDiff: Latent exposure diffusion for HDR generation

    Chao Wang, Zhihao Xia, Thomas Leimkuhler, Karol Myszkowski, and Xuaner Zhang. LEDiff: Latent exposure diffusion for HDR generation. InProc. of Computer Vision and Pattern Recognition, pages 453–464, 2025

    2025

  52. [52]

    An asynchronous Kalman filter for hybrid event cameras

    Ziwei Wang, Yonhon Ng, Cedric Scheerlinck, and Robert Mahony. An asynchronous Kalman filter for hybrid event cameras. InProc. of International Conference on Computer Vision, pages 448–457, 2021

    2021

  53. [53]

    CBAM: Convolutional block attention module

    Sanghyun Woo, Jongchan Park, Joon-Young Lee, and In So Kweon. CBAM: Convolutional block attention module. InProc. of European Conference on Computer Vision, pages 3–19, 2018

    2018

  54. [54]

    HDRFlow: Real-time HDR video reconstruction with large motions

    Gangwei Xu, Yujin Wang, Jinwei Gu, Tianfan Xue, and Xin Yang. HDRFlow: Real-time HDR video reconstruction with large motions. In Proc. of Computer Vision and Pattern Recognition, pages 24851–24860, 2024

    2024

  55. [55]

    Dual-attention-guided network for ghost-free high dynamic range imaging.International Journal of Computer Vision, 130(1):76–94, 2022

    Qingsen Yan, Dong Gong, Javen Qinfeng Shi, Anton Van Den Hengel, Chunhua Shen, Ian Reid, and Yanning Zhang. Dual-attention-guided network for ghost-free high dynamic range imaging.International Journal of Computer Vision, 130(1):76–94, 2022

    2022

  56. [56]

    Real-data-driven 2000 FPS color video from mosaicked chromatic spikes

    Siqi Yang, Zhaojun Huang, Yakun Chang, Bin Fan, Zhaofei Yu, and Boxin Shi. Real-data-driven 2000 FPS color video from mosaicked chromatic spikes. InProc. of European Conference on Computer Vision, pages 305–321, 2024

    2000

  57. [57]

    Adding conditional control to text-to-image diffusion models

    Lvmin Zhang, Anyi Rao, and Maneesh Agrawala. Adding conditional control to text-to-image diffusion models. InProc. of International Conference on Computer Vision, pages 3836–3847, 2023

    2023

  58. [58]

    Unbounded high dynamic range photography using a modulo camera

    Hang Zhao, Boxin Shi, Christy Fernandez-Cull, Sai-Kit Yeung, and Ramesh Raskar. Unbounded high dynamic range photography using a modulo camera. InProc. of International Conference on Computational Photography, pages 1–10, 2015

    2015

  59. [59]

    Polarization guided HDR reconstruction via pixel-wise depolarization.IEEE Transactions on Image Processing, 32:1774–1787, 2023

    Chu Zhou, Yufei Han, Minggui Teng, Jin Han, Si Li, Chao Xu, and Boxin Shi. Polarization guided HDR reconstruction via pixel-wise depolarization.IEEE Transactions on Image Processing, 32:1774–1787, 2023

    2023

  60. [60]

    UnModNet: Learning to unwrap a modulo image for high dynamic range imaging

    Chu Zhou, Hang Zhao, Jin Han, Chang Xu, Chao Xu, Tiejun Huang, and Boxin Shi. UnModNet: Learning to unwrap a modulo image for high dynamic range imaging. InProc. of Advances in Neural Information Processing Systems, 2020

    2020

  61. [61]

    A retina- inspired sampling method for visual texture reconstruction

    Lin Zhu, Siwei Dong, Tiejun Huang, and Yonghong Tian. A retina- inspired sampling method for visual texture reconstruction. InProc. of International Conference on Multimedia and Expo, pages 1432–1437, 2019

    2019

  62. [62]

    Deformable ConvNets v2: More deformable, better results

    Xizhou Zhu, Han Hu, Stephen Lin, and Jifeng Dai. Deformable ConvNets v2: More deformable, better results. InProc. of Computer Vision and Pattern Recognition, pages 9308–9316, 2019. Chu Zhoureceived the B.E. degree from Huazhong University of Science and Technology in 2019 and the Ph.D. degree from School of Intelligence Science and Technology, Peking Univ...

    2019