pith. sign in

arxiv: 2601.05249 · v2 · submitted 2026-01-08 · 💻 cs.CV

RL-AWB: Deep Reinforcement Learning for Auto White Balance Correction in Low-Light Night-time Scenes

Yuan-Kang Lee , Kuan-Lin Chen , Chia-Che Chang , Yu-Lun Liu This is my paper

Pith reviewed 2026-05-16 15:57 UTC · model grok-4.3

classification 💻 cs.CV
keywords auto white balancecolor constancyreinforcement learninglow-light imagingnighttime scenescomputational photographymulti-sensor datasetdeep learning
0
0 comments X

The pith

A reinforcement learning agent learns to tune a statistical white balance estimator per image for better color correction in night scenes.

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

The paper presents RL-AWB as a hybrid system that first runs a custom statistical algorithm to detect salient gray pixels and estimate illumination under low light, then feeds that output as state to a deep reinforcement learning agent. The agent learns a policy to adjust the statistical method's parameters on each input image, copying the iterative tuning process used by human experts. This setup is tested on a newly collected multi-sensor nighttime dataset that enables evaluation across different cameras. The central result is that the learned policy improves color constancy on nighttime images while also performing well on well-lit daytime images, showing generalization beyond the training distribution.

Core claim

By casting auto white balance parameter selection as a reinforcement learning task whose state is supplied by a nighttime-specific statistical illumination estimator, an agent can discover dynamic correction policies that outperform both pure statistical baselines and prior learning-based methods on low-light color constancy while maintaining accuracy on standard illumination.

What carries the argument

The reinforcement learning agent that receives the statistical estimator's output vector as state and selects parameter adjustments as actions to maximize a color accuracy reward.

Load-bearing premise

The statistical nighttime illumination estimator produces a sufficiently stable and informative state that allows the reinforcement learning agent to learn useful parameter policies.

What would settle it

Training the same reinforcement learning agent on raw image pixels without the statistical preprocessing step and showing that it fails to match or exceed the hybrid method's accuracy on the multi-sensor nighttime test set.

Figures

Figures reproduced from arXiv: 2601.05249 by Chia-Che Chang, Kuan-Lin Chen, Yuan-Kang Lee, Yu-Lun Liu.

Figure 1
Figure 1. Figure 1: Our method achieves optimal parameter tuning for automatic white [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Overview of the proposed RL-AWB framework. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Sample images from the proposed LEVI dataset with their corre [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Dataset statistics of the LEVI and NCC datasets. [PITH_FULL_IMAGE:figures/full_fig_p010_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Visual comparison of cross-dataset performance under domain shift. [PITH_FULL_IMAGE:figures/full_fig_p012_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Illustration of the RL-AWB auto-tuning process for representative [PITH_FULL_IMAGE:figures/full_fig_p013_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Failure case analysis. Examples where RL-AWB over-corrects challenging nighttime scenes, resulting in visually degraded outputs. 6 Conclusion This study is the first to apply reinforcement learning to color constancy, demon￾strating that DRL can be effectively used for white balance tuning. Our work makes three contributions: (1) SGP-LRD, a novel statistical algorithm for night￾time color constancy, (2) RL… view at source ↗
Figure 8
Figure 8. Figure 8: Example nighttime scenes from the LEVI dataset. [PITH_FULL_IMAGE:figures/full_fig_p027_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Comparison of cross-sensor performance between our method and [PITH_FULL_IMAGE:figures/full_fig_p031_9.png] view at source ↗
read the original abstract

Nighttime color constancy still remains a challenging problem in computational photography due to low-light noise and complex illumination conditions. We present RL-AWB, a novel framework combining statistical methods with deep reinforcement learning for nighttime white balance. Our method begins with a statistical algorithm tailored for nighttime scenes, integrating salient gray pixel detection with novel illumination estimation. Building on this foundation, we develop the first deep reinforcement learning approach for color constancy that leverages the statistical algorithm as its core, mimicking professional AWB tuning experts by dynamically optimizing parameters for each image. To facilitate cross-sensor evaluation, we introduce the first multi-sensor nighttime dataset. Experiment results show that our method achieves superior generalization capability across low-light and well-illuminated images. Project page: https://ntuneillee.github.io/research/rl-awb/

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 / 2 minor

Summary. The paper proposes RL-AWB, a framework that combines a statistical algorithm (salient gray pixel detection plus novel illumination estimation tailored for nighttime scenes) with deep reinforcement learning to dynamically optimize auto white balance parameters, mimicking expert tuning. It introduces the first multi-sensor nighttime dataset for cross-sensor evaluation and claims that the resulting method achieves superior generalization across both low-light nighttime and well-illuminated images.

Significance. If the empirical claims hold after proper validation, the work would be significant as the first deep RL formulation for color constancy, the introduction of a new multi-sensor nighttime dataset, and a hybrid statistical-RL approach that could improve robustness in challenging illumination regimes. The explicit goal of cross-regime generalization is a valuable direction for computational photography.

major comments (2)
  1. [Experiments] Experiments section: the headline claim of superior generalization across low-light and well-illuminated images is unsupported because no quantitative metrics, baselines, ablation studies, or error analysis are supplied in the manuscript; without these the central empirical assertion cannot be verified.
  2. [Method] Method section: the RL agent uses the output of the nighttime statistical algorithm as its state representation; no ablation isolates whether the learned policy actually improves upon the statistical baseline itself on well-illuminated test images, so the generalization result could be an artifact of dataset composition rather than genuine cross-regime improvement.
minor comments (2)
  1. [Method] The abstract states that the statistical algorithm is 'leveraged as its core' but does not provide the explicit state-action formulation or reward function used by the RL agent; these should be given in equations.
  2. [Dataset] Dataset description: the multi-sensor nighttime dataset is introduced but no details on sensor characteristics, number of images per sensor, or train/test splits are provided.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive comments. We agree that the current experiments section requires substantial strengthening to support the generalization claims, and we will revise the manuscript to include the requested quantitative evaluations and ablations.

read point-by-point responses

  1. Referee: [Experiments] Experiments section: the headline claim of superior generalization across low-light and well-illuminated images is unsupported because no quantitative metrics, baselines, ablation studies, or error analysis are supplied in the manuscript; without these the central empirical assertion cannot be verified.

    Authors: We acknowledge this limitation in the submitted manuscript. In the revised version we will expand the Experiments section with quantitative results using standard metrics (angular error, PSNR, SSIM) on both the new multi-sensor nighttime dataset and well-illuminated images drawn from established benchmarks. We will report comparisons against multiple statistical and learning-based baselines, include full ablation tables isolating each component, and provide error analysis (e.g., per-scene failure cases and cross-sensor variance) to substantiate the generalization claims. revision: yes

  2. Referee: [Method] Method section: the RL agent uses the output of the nighttime statistical algorithm as its state representation; no ablation isolates whether the learned policy actually improves upon the statistical baseline itself on well-illuminated test images, so the generalization result could be an artifact of dataset composition rather than genuine cross-regime improvement.

    Authors: We agree that isolating the RL policy's contribution on well-illuminated images is necessary. The revised manuscript will add a dedicated ablation that directly compares the full RL-AWB pipeline against the statistical baseline alone (with RL disabled) on well-illuminated test images. This will quantify any additional improvement provided by the learned policy and rule out dataset-composition artifacts. revision: yes

Circularity Check

0 steps flagged

No circularity detected; RL framework uses statistical baseline as state input without reducing performance claims to fitted quantities or self-definitions.

full rationale

The paper describes a statistical algorithm (salient gray pixel detection + illumination estimation) that supplies the state for an RL agent which then optimizes AWB parameters. The central claim of superior generalization is presented as an empirical outcome on a newly introduced multi-sensor nighttime dataset, not as a mathematical identity or prediction forced by fitting parameters to the evaluation data itself. No equations, uniqueness theorems, or ansatzes are shown that would make the RL output equivalent to the statistical input by construction. Self-citations, if present in the full text, are not load-bearing for the derivation. This is a standard empirical ML contribution whose validity rests on experimental results rather than tautological reduction.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only information prevents identification of specific free parameters, axioms, or invented entities; no numerical constants, unproven assumptions, or new postulated objects are described.

pith-pipeline@v0.9.0 · 5442 in / 968 out tokens · 31290 ms · 2026-05-16T15:57:01.326625+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

125 extracted references · 125 canonical work pages · 5 internal anchors

  1. [1]

    In: Proceedings of the IEEE/CVF International Con- ference on Computer Vision

    Afifi, M., Barron, J.T., LeGendre, C., Tsai, Y.T., Bleibel, F.: Cross-camera con- volutional color constancy. In: Proceedings of the IEEE/CVF International Con- ference on Computer Vision. pp. 1981–1990 (2021)

    work page 1981

  2. [2]

    arXiv preprint arXiv:1912.06888 (2019)

    Afifi, M., Brown, M.S.: Sensor-independent illumination estimation for dnn mod- els. arXiv preprint arXiv:1912.06888 (2019)

    work page arXiv 1912

  3. [3]

    In: Proceedings of the IEEE/CVF Conference on computer vision and pattern recognition

    Afifi, M., Brown, M.S.: Deep white-balance editing. In: Proceedings of the IEEE/CVF Conference on computer vision and pattern recognition. pp. 1397– 1406 (2020)

    work page 2020

  4. [4]

    In: Proceedings of the IEEE/CVF Winter Conference on Ap- plications of Computer Vision

    Afifi, M., Brubaker, M.A., Brown, M.S.: Auto white-balance correction for mixed- illuminant scenes. In: Proceedings of the IEEE/CVF Winter Conference on Ap- plications of Computer Vision. pp. 1210–1219 (2022)

    work page 2022

  5. [5]

    In: European Conference on Computer Vision

    Afifi, M., Hu, Z., Liang, L.: Optimizing illuminant estimation in dual-exposure hdr imaging. In: European Conference on Computer Vision. pp. 202–219. Springer (2024)

    work page 2024

  6. [6]

    Advances in neural information processing systems30(2017)

    Andrychowicz, M., Wolski, F., Ray, A., Schneider, J., Fong, R., Welinder, P., McGrew, B., Tobin, J., Pieter Abbeel, O., Zaremba, W.: Hindsight experience replay. Advances in neural information processing systems30(2017)

    work page 2017

  7. [7]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Bai, M., Urtasun, R.: Deep watershed transform for instance segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 5221–5229 (2017)

    work page 2017

  8. [8]

    In: Advances in Data-driven Computing and Intelli- gent Systems: Selected Papers from ADCIS 2022, Volume 2, pp

    Bajaj, C., Yang, Y., Wang, Y.: Reinforcement learning of self-enhancing camera image and signal processing. In: Advances in Data-driven Computing and Intelli- gent Systems: Selected Papers from ADCIS 2022, Volume 2, pp. 281–303. Springer (2023)

    work page 2022

  9. [9]

    i: Methodology and experiments with synthesized data

    Barnard, K., Cardei, V., Funt, B.: A comparison of computational color con- stancy algorithms. i: Methodology and experiments with synthesized data. IEEE transactions on Image Processing11(9), 972–984 (2002)

    work page 2002

  10. [10]

    In: Proceedings of the IEEE Inter- national Conference on Computer Vision

    Barron, J.T.: Convolutional color constancy. In: Proceedings of the IEEE Inter- national Conference on Computer Vision. pp. 379–387 (2015)

    work page 2015

  11. [11]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Barron, J.T., Tsai, Y.T.: Fast fourier color constancy. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 886–894 (2017)

    work page 2017

  12. [12]

    In: Proceedings of the 26th annual international conference on machine learning

    Bengio, Y., Louradour, J., Collobert, R., Weston, J.: Curriculum learning. In: Proceedings of the 26th annual international conference on machine learning. pp. 41–48 (2009)

    work page 2009

  13. [13]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Bianco, S., Cusano, C.: Quasi-unsupervised color constancy. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 12212–12221 (2019) 16 H.-S. Shiu et al

    work page 2019

  14. [14]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition workshops

    Bianco, S., Cusano, C., Schettini, R.: Color constancy using cnns. In: Proceedings of the IEEE conference on computer vision and pattern recognition workshops. pp. 81–89 (2015)

    work page 2015

  15. [15]

    IEEE Transactions on Image Processing26(9), 4347–4362 (2017)

    Bianco,S.,Cusano,C.,Schettini,R.:Singleandmultipleilluminantestimationus- ing convolutional neural networks. IEEE Transactions on Image Processing26(9), 4347–4362 (2017)

    work page 2017

  16. [16]

    Journal of the optical Society of America A14(7), 1393–1411 (1997)

    Brainard, D.H., Freeman, W.T.: Bayesian color constancy. Journal of the optical Society of America A14(7), 1393–1411 (1997)

    work page 1997

  17. [17]

    Journal of the Franklin institute310(1), 1–26 (1980)

    Buchsbaum, G.: A spatial processor model for object colour perception. Journal of the Franklin institute310(1), 1–26 (1980)

    work page 1980

  18. [18]

    Pattern Recognition160, 111175 (2025)

    Buzzelli, M., Bianco, S.: Uncertainty estimation in color constancy. Pattern Recognition160, 111175 (2025)

    work page 2025

  19. [19]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Cai, Y., Bian, H., Lin, J., Wang, H., Timofte, R., Zhang, Y.: Retinexformer: One- stage retinex-based transformer for low-light image enhancement. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 12504–12513 (2023)

    work page 2023

  20. [20]

    IEEE Transactions on Pattern Analysis and Machine Intelligence34(8), 1509–1519 (2011)

    Chakrabarti, A., Hirakawa, K., Zickler, T.: Color constancy with spatio-spectral statistics. IEEE Transactions on Pattern Analysis and Machine Intelligence34(8), 1509–1519 (2011)

    work page 2011

  21. [21]

    In: Proceed- ings of the Computer Vision and Pattern Recognition Conference

    Chang, C.W., Fan, C.D., Chang, C.C., Lo, Y.C., Tseng, Y.C., Huang, J.L., Liu, Y.L.: Gcc: Generative color constancy via diffusing a color checker. In: Proceed- ings of the Computer Vision and Pattern Recognition Conference. pp. 10868– 10878 (2025)

    work page 2025

  22. [22]

    In: European Conference on Com- puter Vision

    Chang, K.C., Wang, R., Lin, H.J., Liu, Y.L., Chen, C.P., Chang, Y.L., Chen, H.T.: Learning camera-aware noise models. In: European Conference on Com- puter Vision. pp. 343–358. Springer (2020)

    work page 2020

  23. [23]

    In: Pro- ceedings of the IEEE conference on computer vision and pattern recognition

    Chen, C., Chen, Q., Xu, J., Koltun, V.: Learning to see in the dark. In: Pro- ceedings of the IEEE conference on computer vision and pattern recognition. pp. 3291–3300 (2018)

    work page 2018

  24. [24]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Chen, L., Zhang, Y., Song, Y., Shan, Y., Liu, L.: Improved test-time adapta- tion for domain generalization. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 24172–24182 (2023)

    work page 2023

  25. [25]

    In: Proceedings of the IEEE/CVF International Conference on Com- puter Vision

    Chen, S.K., Yen, H.L., Liu, Y.L., Chen, M.H., Hu, H.N., Peng, W.H., Lin, Y.Y.: Learning continuous exposure value representations for single-image hdr recon- struction. In: Proceedings of the IEEE/CVF International Conference on Com- puter Vision. pp. 12990–13000 (2023)

    work page 2023

  26. [26]

    Journal of the Optical Society of America A 41(3), 476–488 (2024)

    Cheng, C., Yang, K.F., Wan, X.M., Chan, L.L.H., Li, Y.J.: Nighttime color con- stancy using robust gray pixels. Journal of the Optical Society of America A 41(3), 476–488 (2024)

    work page 2024

  27. [27]

    Journal of the Optical Society of America A31(5), 1049–1058 (2014)

    Cheng, D., Prasad, D.K., Brown, M.S.: Illuminant estimation for color constancy: why spatial-domain methods work and the role of the color distribution. Journal of the Optical Society of America A31(5), 1049–1058 (2014)

    work page 2014

  28. [28]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Cheng, D., Price, B., Cohen, S., Brown, M.S.: Effective learning-based illumi- nant estimation using simple features. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 1000–1008 (2015)

    work page 2015

  29. [29]

    In: Proceedings of the AAAI Conference on Artificial Intelligence

    Conde, M.V., McDonagh, S., Maggioni, M., Leonardis, A., Pérez-Pellitero, E.: Model-based image signal processors via learnable dictionaries. In: Proceedings of the AAAI Conference on Artificial Intelligence. vol. 36, pp. 481–489 (2022)

    work page 2022

  30. [30]

    arXiv preprint (2022) arXiv:2203.11068

    Cun, X., Wang, Z., Pun, C.M., Liu, J., Zhou, W., Jia, X., Li, H.: Learning en- riched illuminants for cross and single sensor color constancy. arXiv preprint arXiv:2203.11068 (2022) RL-AWB 17

    work page arXiv 2022

  31. [31]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Das, P., Liu, Y., Karaoglu, S., Gevers, T.: Generative models for multi- illumination color constancy. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 1194–1203 (2021)

    work page 2021

  32. [32]

    Machine Vision and Applications20(5), 283–301 (2009)

    Ebner, M.: Color constancy based on local space average color. Machine Vision and Applications20(5), 283–301 (2009)

    work page 2009

  33. [33]

    IEEE Transactions on Image Processing9(10), 1774–1783 (2000)

    Finlayson, G., Hordley, S.: Improving gamut mapping color constancy. IEEE Transactions on Image Processing9(10), 1774–1783 (2000)

    work page 2000

  34. [34]

    In: Proceedings of the IEEE International Conference on Computer Vision

    Finlayson, G.D.: Corrected-moment illuminant estimation. In: Proceedings of the IEEE International Conference on Computer Vision. pp. 1904–1911 (2013)

    work page 1904

  35. [35]

    In: Color and imaging conference

    Finlayson, G.D., Trezzi, E.: Shades of gray and colour constancy. In: Color and imaging conference. vol. 12, pp. 37–41. Society of Imaging Science and Technology (2004)

    work page 2004

  36. [36]

    In: Conference on robot learning

    Florensa, C., Held, D., Wulfmeier, M., Zhang, M., Abbeel, P.: Reverse curriculum generation for reinforcement learning. In: Conference on robot learning. pp. 482–

    work page

  37. [37]

    International Journal of Computer Vision5(1), 5–35 (1990)

    Forsyth, D.A.: A novel algorithm for color constancy. International Journal of Computer Vision5(1), 5–35 (1990)

    work page 1990

  38. [38]

    IEEE Transactions on Multimedia 22(7), 1704–1719 (2019)

    Furuta, R., Inoue, N., Yamasaki, T.: Pixelrl: Fully convolutional network with reinforcement learning for image processing. IEEE Transactions on Multimedia 22(7), 1704–1719 (2019)

    work page 2019

  39. [39]

    In: 2008 IEEE Conference on Computer Vision and Pattern Recognition

    Gehler, P.V., Rother, C., Blake, A., Minka, T., Sharp, T.: Bayesian color con- stancy revisited. In: 2008 IEEE Conference on Computer Vision and Pattern Recognition. pp. 1–8. IEEE (2008)

    work page 2008

  40. [40]

    Gijsenij, A., Gevers, T., Van De Weijer, J.: Generalized gamut mapping using imagederivativestructuresforcolorconstancy.InternationalJournalofComputer Vision86(2), 127–139 (2010)

    work page 2010

  41. [41]

    IEEE transactions on image processing20(9), 2475–2489 (2011)

    Gijsenij,A.,Gevers,T.,VanDeWeijer,J.:Computationalcolorconstancy:Survey and experiments. IEEE transactions on image processing20(9), 2475–2489 (2011)

    work page 2011

  42. [42]

    IEEE Transactions on Pattern Analysis and Machine Intelligence34(5), 918–929 (2011)

    Gijsenij, A., Gevers, T., Van De Weijer, J.: Improving color constancy by pho- tometric edge weighting. IEEE Transactions on Pattern Analysis and Machine Intelligence34(5), 918–929 (2011)

    work page 2011

  43. [43]

    In: international conference on machine learning

    Graves,A.,Bellemare,M.G.,Menick,J.,Munos,R.,Kavukcuoglu,K.:Automated curriculum learning for neural networks. In: international conference on machine learning. pp. 1311–1320. Pmlr (2017)

    work page 2017

  44. [44]

    In: Pro- ceedings of the 27th international conference on international conference on ma- chine learning

    Gregor, K., LeCun, Y.: Learning fast approximations of sparse coding. In: Pro- ceedings of the 27th international conference on international conference on ma- chine learning. pp. 399–406 (2010)

    work page 2010

  45. [45]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Guo, C., Li, C., Guo, J., Loy, C.C., Hou, J., Kwong, S., Cong, R.: Zero-reference deep curve estimation for low-light image enhancement. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 1780– 1789 (2020)

    work page 2020

  46. [46]

    International Conference on Machine Learning (ICML) (2018)

    Haarnoja, T., Zhou, A., Abbeel, P., Levine, S.: Soft actor-critic: Off-policy max- imum entropy deep reinforcement learning with a stochastic actor. International Conference on Machine Learning (ICML) (2018)

    work page 2018

  47. [47]

    Soft Actor-Critic Algorithms and Applications

    Haarnoja, T., Zhou, A., Hartikainen, K., Tucker, G., Ha, S., Tan, J., Kumar, V., Zhu, H., Gupta, A., Abbeel, P., et al.: Soft actor-critic algorithms and applica- tions. arXiv preprint arXiv:1812.05905 (2018)

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  48. [48]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Hernandez-Juarez, D., Parisot, S., Busam, B., Leonardis, A., Slabaugh, G., Mc- Donagh, S.: A multi-hypothesis approach to color constancy. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 2270– 2280 (2020) 18 H.-S. Shiu et al

    work page 2020

  49. [49]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Hu, Y., Wang, B., Lin, S.: Fc4: Fully convolutional color constancy with confidence-weighted pooling. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 4085–4094 (2017)

    work page 2017

  50. [50]

    ACM Transactions on Graphics (ToG)42(6), 1– 14 (2023)

    Jiang, H., Luo, A., Fan, H., Han, S., Liu, S.: Low-light image enhancement with wavelet-based diffusion models. ACM Transactions on Graphics (ToG)42(6), 1– 14 (2023)

    work page 2023

  51. [51]

    In: Proceedings of the AAAI Conference on Artificial Intelligence

    Jiang, L., Meng, D., Zhao, Q., Shan, S., Hauptmann, A.: Self-paced curriculum learning. In: Proceedings of the AAAI Conference on Artificial Intelligence. vol. 29 (2015)

    work page 2015

  52. [52]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Karmanov, A., Guan, D., Lu, S., El Saddik, A., Xing, E.: Efficient test-time adap- tation of vision-language models. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 14162–14171 (2024)

    work page 2024

  53. [53]

    arXiv preprint arXiv:2504.07959 (2025)

    Kim, D., Afifi, M., Kim, D., Brown, M.S., Kim, S.J.: Ccmnet: Leveraging cali- brated color correction matrices for cross-camera color constancy. arXiv preprint arXiv:2504.07959 (2025)

    work page arXiv 2025

  54. [54]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Kim, D., Kim, J., Nam, S., Lee, D., Lee, Y., Kang, N., Lee, H.E., Yoo, B., Han, J.J., Kim, S.J.: Large scale multi-illuminant (lsmi) dataset for developing white balance algorithm under mixed illumination. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 2410–2419 (2021)

    work page 2021

  55. [55]

    In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision

    Kınlı, F., Yılmaz, D., Özcan, B., Kıraç, F.: Modeling the lighting in scenes as style for auto white-balance correction. In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. pp. 4903–4913 (2023)

    work page 2023

  56. [56]

    Advances in Neural Information Processing Systems33, 9216–9227 (2020)

    Klink, P., D’Eramo, C., Peters, J.R., Pajarinen, J.: Self-paced deep reinforce- ment learning. Advances in Neural Information Processing Systems33, 9216–9227 (2020)

    work page 2020

  57. [57]

    International Journal of Computer Vision132(2), 287–299 (2024)

    Koskinen, S., Acar, E., Kämäräinen, J.K.: Single pixel spectral color constancy. International Journal of Computer Vision132(2), 287–299 (2024)

    work page 2024

  58. [58]

    In: Proceedings of the AAAI confer- ence on artificial intelligence

    Kosugi, S., Yamasaki, T.: Unpaired image enhancement featuring reinforcement- learning-controlled image editing software. In: Proceedings of the AAAI confer- ence on artificial intelligence. vol. 34, pp. 11296–11303 (2020)

    work page 2020

  59. [59]

    Advances in neural information processing systems23(2010)

    Kumar, M., Packer, B., Koller, D.: Self-paced learning for latent variable models. Advances in neural information processing systems23(2010)

    work page 2010

  60. [60]

    IEEE Transactions on Image Processing29, 7722–7734 (2020)

    Laakom, F., Passalis, N., Raitoharju, J., Nikkanen, J., Tefas, A., Iosifidis, A., Gabbouj, M.: Bag of color features for color constancy. IEEE Transactions on Image Processing29, 7722–7734 (2020)

    work page 2020

  61. [61]

    Journal of the Optical society of America61(1), 1–11 (1971)

    Land, E.H., McCann, J.J.: Lightness and retinex theory. Journal of the Optical society of America61(1), 1–11 (1971)

    work page 1971

  62. [62]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Lee, K., Shin, U., Lee, B.U.: Learning to control camera exposure via reinforce- ment learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 2975–2983 (2024)

    work page 2024

  63. [63]

    Embedding fourier for ultra-high-definition low-light im- age enhancement.arXiv preprint arXiv:2302.11831, 2023

    Li, C., Guo, C.L., Zhou, M., Liang, Z., Zhou, S., Feng, R., Loy, C.C.: Embed- ding fourier for ultra-high-definition low-light image enhancement. arXiv preprint arXiv:2302.11831 (2023)

    work page arXiv 2023

  64. [64]

    IEEE transactions on pattern analysis and machine intelligence44(12), 9396–9416 (2021)

    Li, C., Guo, C., Han, L., Jiang, J., Cheng, M.M., Gu, J., Loy, C.C.: Low-light image and video enhancement using deep learning: A survey. IEEE transactions on pattern analysis and machine intelligence44(12), 9396–9416 (2021)

    work page 2021

  65. [65]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Li, J., Chen, C., Huang, W., Lang, Z., Song, F., Yan, Y., Xiong, Z.: Learning steer- able function for efficient image resampling. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 5866–5875 (2023) RL-AWB 19

    work page 2023

  66. [66]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition

    Li, S., Tan, R.T.: Nightcc: Nighttime color constancy via adaptive channel mask- ing. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition. pp. 25522–25531 (2024)

    work page 2024

  67. [67]

    CoRR (2022)

    Li, S., Wang, J., Brown, M.S., Tan, R.T.: Transcc: Transformer-based multiple illuminant color constancy using multitask learning. CoRR (2022)

    work page 2022

  68. [68]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Li,Z.,Yi,S.,Ma,Z.:Renderingnighttimeimageviacascadedcolorandbrightness compensation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 897–905 (2022)

    work page 2022

  69. [69]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Liu, X., Wu, Z., Li, A., Vasluianu, F.A., Zhang, Y., Gu, S., Zhang, L., Zhu, C., Timofte, R., Jin, Z., et al.: Ntire 2024 challenge on low light image enhancement: Methods and results. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 6571–6594 (2024)

    work page 2024

  70. [70]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Liu, Y.L., Lai, W.S., Chen, Y.S., Kao, Y.L., Yang, M.H., Chuang, Y.Y., Huang, J.B.: Single-image hdr reconstruction by learning to reverse the camera pipeline. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 1651–1660 (2020)

    work page 2020

  71. [71]

    Liu, Y., Kothari, P., Van Delft, B., Bellot-Gurlet, B., Mordan, T., Alahi, A.: Ttt++: When does self-supervised test-time training fail or thrive? Advances in Neural Information Processing Systems34, 21808–21820 (2021)

    work page 2021

  72. [72]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Lo, Y.C., Chang, C.C., Chiu, H.C., Huang, Y.H., Chen, C.P., Chang, Y.L., Jou, K.: Clcc: Contrastive learning for color constancy. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 8053– 8063 (2021)

    work page 2021

  73. [73]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Ma,L.,Ma,T.,Liu,R.,Fan,X.,Luo,Z.:Towardfast,flexible,androbustlow-light image enhancement. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 5637–5646 (2022)

    work page 2022

  74. [74]

    IEEE transactions on neural networks and learning systems31(9), 3732– 3740 (2019)

    Matiisen, T., Oliver, A., Cohen, T., Schulman, J.: Teacher–student curriculum learning. IEEE transactions on neural networks and learning systems31(9), 3732– 3740 (2019)

    work page 2019

  75. [75]

    Formulating Camera-Adaptive Color Constancy as a Few-shot Meta-Learning Problem

    McDonagh, S., Parisot, S., Zhou, F., Zhang, X., Leonardis, A., Li, Z., Slabaugh, G.: Formulating camera-adaptive color constancy as a few-shot meta-learning problem. arXiv preprint arXiv:1811.11788 (2018)

    work page internal anchor Pith review Pith/arXiv arXiv 2018

  76. [76]

    IEEE Signal Processing Magazine38(2), 18–44 (2021)

    Monga, V., Li, Y., Eldar, Y.C.: Algorithm unrolling: Interpretable, efficient deep learning for signal and image processing. IEEE Signal Processing Magazine38(2), 18–44 (2021)

    work page 2021

  77. [77]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Mosleh, A., Sharma, A., Onzon, E., Mannan, F., Robidoux, N., Heide, F.: Hardware-in-the-loop end-to-end optimization of camera image processing pipelines. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 7529–7538 (2020)

    work page 2020

  78. [78]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Mou, C., Wang, Q., Zhang, J.: Deep generalized unfolding networks for image restoration. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 17399–17410 (2022)

    work page 2022

  79. [79]

    Journal of Machine Learning Research21(181), 1–50 (2020)

    Narvekar, S., Peng, B., Leonetti, M., Sinapov, J., Taylor, M.E., Stone, P.: Cur- riculum learning for reinforcement learning domains: A framework and survey. Journal of Machine Learning Research21(181), 1–50 (2020)

    work page 2020

  80. [80]

    Pattern Recognition61, 405–416 (2017)

    Oh, S.W., Kim, S.J.: Approaching the computational color constancy as a classi- fication problem through deep learning. Pattern Recognition61, 405–416 (2017)

    work page 2017

Showing first 80 references.