Pith. sign in

REVIEW 4 major objections 5 minor 30 references

Night-to-Day Translation via Illumination Degradation Disentanglement

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper claims that nighttime images translate better to daytime when light effects are first separated from well-lit regions using a Kubelka-Munk color invariant.

desk verdict Solid, empirically strong Night2Day method; the physics story needs the missing proof and a robustness check before the SOTA claim is bulletproof. read the letter →

arxiv 2411.14504 v1 pith:NLY66TJV submitted 2024-11-21 cs.CV

Guanzhou Lan , Yuqi Yang , Zhigang Wang , Dong Wang , Bin Zhao , Xuelong Li This is my paper · ORCID
classification cs.CV
keywords night-to-daytranslationunpairedimageilluminationdegradationdisentanglementKubelka-Munkcolorinvariantcontrastivelearningsemanticsegmentationautonomousdrivinglow-lightenhancement
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper proposes a way to translate nighttime images into day-like images without paired training data by first separating the image into four degradation types: darkness, well-lit regions, high-light regions, and light effects such as flare and glow. The central claim is that treating light effects as a distinct degradation, rather than folding them into well-lit regions, is what lets a GAN-based translator preserve structure and produce realistic daytime output. On the Alderley rainy-night dataset the method improves the Frechet Inception Distance (FID) by 10.3 and the SIFT keypoint-matching score by 4.52 over the prior best, and on BDD100K it improves FID by 5.4 and raises nighttime semantic segmentation mIoU from 15.63 to 21.58 with a Cityscapes-pretrained model. The reason this matters is that downstream tasks such as localization and segmentation inherit the gain without extra nighttime annotations.

What carries the argument

The load-bearing object is the Kubelka-Munk color invariant $N_{\lambda^m x^n}$, a differential expression built from the observed spectrum $E(\lambda,x)$ and its derivatives that, under a local uniformity assumption, reduces to a function of the illumination $e(\lambda,x)$ alone. The paper's simplified photometric model splits the night image into well-lit regions where reflected light is $e(\lambda,x)R(\lambda)C(x)$ and light-effect regions where reflected light is pure $e(\lambda,x)$; because $N$ depends only on $e$, it isolates light effects from well-lit areas that share similar intensity. In practice $N$ is computed via the Gaussian color model and spatial derivatives, thresholded with ReLU and normalized to produce the light-effect mask $M_{le}$, and that mask drives both the disentanglement map and the contrastive sampling and reweighting.

What would settle it

Take a real night photograph with a known object-level specular highlight, for example a glossy car hood reflecting a streetlight, and compute the light-effect mask $M_{le}$ from Eq. (9); if the mask either fails to cover the highlight or bleeds into the surrounding matte well-lit surface while the disentanglement maps flip, the piecewise Fresnel assumption in Eq. (3) is violated and the physical prior no longer carries the claimed gain.

Watch

Extended reading notes

Core claim

The paper's central discovery is that the failure of previous unpaired night-to-day translators comes from treating all illuminated nighttime regions uniformly, and that a physically grounded invariant can fix this. Using a photometric model derived from Kubelka-Munk theory, the authors derive a color invariant $N$ that responds only to the illumination spectrum $e(\lambda,x)$, not to material reflectance; they show this invariant acts as a light-effects detector, letting them carve the nighttime image into darkness, well-lit, high-light, and light-effect masks. Those masks then guide a contrastive learning objective in which anchors and negatives are sampled within each degradation class and reweighted by an optimal transport plan, turning the similarity matrix into a block-diagonal one. The paper argues, and demonstrates on two datasets, that this degradation-aware contrastive loss preserves semantic structure across all regions and yields the best FID, LPIPS, SIFT, and mIoU numbers among the methods compared.

Load-bearing premise

The method assumes that every pixel in the mixed light-effect and well-lit region is either pure illumination or pure reflectance, with no partial specularity, so the Fresnel coefficient is always 0 or 1; if real glow, flare, or shiny surfaces blend the two, the invariant no longer isolates illumination and the masks are corrupted.

Editorial extensions

If this is right

  • Night-to-day translation can be improved without paired data or human annotations by using physical priors to separate degradation types.
  • Downstream nighttime vision tasks, including semantic segmentation and keypoint-based localization, improve when the translation model is degradation-aware; the paper's BDD100K segmentation gain comes from translating the image and then running a daytime-pretrained model.
  • The disentanglement map itself, not just the final image, could serve as a supervisory signal or an attention mechanism for other nighttime perception tasks.
  • The contrastive reweighting result implies that hard negatives within the same degradation class matter more than easy cross-class negatives, so similarity matrices should be block-diagonalized by degradation type.
  • The ablations indicate that illuminance clustering alone, whether into three or four clusters, is not enough; the additional physics-derived invariant is what separates light effects from well-lit regions.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The same Kubelka-Munk invariant could be reused as a standalone light-effect detector in other night-vision pipelines, such as low-light enhancement or autonomous driving perception, where glare and flare are usually treated as noise rather than as a separate domain.
  • Because the method needs only a single night image and no daytime pair, it could be extended to video by propagating the disentanglement masks across frames, letting temporal consistency constraints replace some of the contrastive sampling.
  • A testable extension is to replace the hard piecewise assumption in Eq. (3) with a continuous Fresnel coefficient and let the mask encode soft membership; if performance holds, the physical prior is more robust than the simplified model suggests.
  • The optimal-transport reweighting is agnostic to the source of the masks, so feeding masks from a learned segmenter or from a different invariant could improve performance further.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes N2D3, an unpaired night-to-day image translation method that separates nighttime degradations into darkness, well-lit, high-light, and light-effect regions. It first estimates an illuminance map and applies k-means for the first three regions, then uses a Kubelka-Munk-based color invariant N to isolate light effects from well-lit regions. The resulting masks guide a degradation-aware contrastive learning module with disentanglement-guided sampling and optimal-transport reweighting. Experiments on Alderley and BDD100K report state-of-the-art FID, SIFT, LPIPS, and downstream semantic segmentation mIoU results, and ablations support the contributions of the two main modules.

Significance. If the reported results hold, the paper is a meaningful empirical contribution to unpaired night-to-day translation and to nighttime perception more broadly. The core idea of using a physical invariant to disentangle light effects before contrastive sampling is plausible, and the reported margins over prior methods are large. I also credit the authors for avoiding circular evaluation: the physical invariant is adopted from published color-invariance work, and the contrastive loss is trained on source and generated image patches rather than on FID, SIFT, or mIoU targets. However, the central claim rests on several load-bearing assumptions—especially the binary Fresnel split in Eq. (3)—and the lack of uncertainty quantification in the main table makes the superiority claim less secure than the narrative suggests.

major comments (4)
  1. [§3.1, Eqs. (1)–(3)] The disentanglement pipeline rests on the binary Fresnel split in Eq. (3), where every pixel is assigned either rho_f(x)=1 (light effects, E=e) or rho_f(x)=0 (well-lit, E=eRC). Equation (1), however, contains the term (1 - rho_f)^2, so for intermediate rho_f the derivative expressions in Eqs. (4)–(6) include rho_f, rho_f_x, and cross terms. In that regime N no longer isolates illumination; it also responds to material geometry and partial specularity, and the ReLU-thresholded mask Mle in Eq. (9) is miscalibrated. Since Mle drives both sampling and reweighting in Section 3.3, this piecewise assumption is load-bearing for the SOTA claim, not a minor technical detail. Please provide a sensitivity analysis for partial specularity, glows, wet-road highlights, and clipped highlights, or validate Mle against manually annotated light-effect regions.
  2. [§3.1, Corollary 1] The proof of Corollary 1 is deferred to a supplementary material that is not included in this submission. Because this corollary is the theoretical justification for using N as a light-effect detector, the claim that N depends only on e(lambda,x) is currently asserted rather than demonstrated. Please include a self-contained proof in the paper, or cite a public source where this exact result is established.
  3. [§4.2–4.3, Table 1] All headline comparisons in Table 1 are reported as point estimates with no error bars, number of seeds, or significance tests. The claimed improvements over the previous best are 10.3 FID on Alderley and 5.4 FID on BDD100K, and GAN training typically has nontrivial run-to-run variance. Please report mean ± std over at least three random seeds for the main metrics and, if possible, a paired significance test for the FID/SIFT/mIoU comparisons.
  4. [§3.1–3.2, Eqs. (7)–(8)] The Gaussian color model matrix in Eq. (7) and the derivative scale sigma in Eq. (8) are fixed without justification or sensitivity analysis. Since the computed invariant N, and therefore Mle, may depend strongly on sigma and on the particular color transformation, please add an ablation over sigma, document the origin of the matrix coefficients, and show that the final FID/SIFT/mIoU results are stable across these choices.
minor comments (5)
  1. [§4.4, Tables 2 and 3] The check-mark symbols in Tables 2 and 3 are not decoded in the captions or legend; please add an explicit legend so readers can identify which rows activate the sampling, reweighting, illuminance map, and physical invariant components.
  2. [§4.4] The sentence stating that degradation-aware sampling causes a 'noticeable decrease in FID' is ambiguous, because the reported FID value decreases (i.e., improves) relative to baseline; please rephrase to 'a substantial reduction in FID'.
  3. [§1 and Abstract] The acronym N2D3 is expanded inconsistently: the abstract uses 'Night-to-Day via Degradation Disentanglement' while the introduction uses 'Night to Day via Degradation Disentanglement' without the hyphen; please use one consistent expansion.
  4. [§3.2, Eq. (9)] The normalization terms µ(N) and σ(N) in Eq. (9) are not defined; please specify whether these are global statistics over the whole dataset or per-image statistics.
  5. [§4.1] Implementation details such as image resolution, batch size, learning rate, number of training iterations, number of sampling layers S, contrastive temperature tau, and the value of sigma used in the experiments are omitted; adding them would substantially improve reproducibility.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the physical invariant is imported from external work, and the reported evaluation metrics are not optimized or defined in terms of the method's outputs.

full rationale

I examined the derivation chain in Sections 3.1–3.4 and the experiments in Section 4. The photometric model in Eq. (1) is attributed to Kubelka-Munk theory via the external reference Geusebroek et al. [5], not to the authors' prior work. The piecewise simplification in Eq. (3) is a stated modeling assumption, and the color invariant in Eq. (4) is presented as Corollary 1 with a proof deferred to the supplementary material; while the missing proof is a completeness concern, it is not circular because the invariant formula is not derived from the evaluation metrics. The masks in Eqs. (8)–(9) are computed from input images via the invariant and k-means, and the contrastive loss in Eqs. (10)–(13) is trained on source and generated patches, not on FID, SIFT, LPIPS, or mIoU values. The reported improvements on Alderley and BDD100K are external benchmark measurements, and no parameter is fitted to those targets. The ablation studies compare variants of the proposed components, which is consistent with empirical evaluation rather than circular reasoning. I found no step where a claimed prediction is equivalent by construction to an input, no load-bearing self-citation, and no fitted parameter renamed as a prediction.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central contribution rests on a small set of domain assumptions inherited from color-invariance theory and on several unspecified hyperparameters. No new physical entities are introduced.

free parameters (3)
  • k-means cluster count = 3
    Manually chosen to partition darkness, well-lit, and high-light regions; affects all downstream masks and the contrastive sampling.
  • Gaussian derivative scale sigma = unspecified
    Appears in Eq. 8 for computing Ex; no value or selection criterion is given, yet it controls the invariant N and therefore the light-effects mask Mle.
  • contrastive temperature tau and encoder layers S = unspecified
    Used in Eq. 10-13; without values the exact contrastive objective is underdetermined and reproducibility is limited.
assumptions (5)
  • domain assumption Kubelka-Munk photometric model Eq. 1 describes nighttime pixel spectra
    Basis of the physical prior; external theory applied to night images, cited as [5].
  • domain assumption Materials are locally uniform and homogeneous, so R_inf(lambda,x)=R(lambda)C(x) (Eq. 2)
    Needed to separate reflectance from illumination in Section 3.1.
  • ad hoc to paper Fresnel reflectance rho_f is approximately 0 in well-lit regions and 1 in light effects, giving Eq. 3
    Piecewise simplification that makes the invariant a light-effect detector; no empirical validation of this dichotomy is provided.
  • domain assumption Corollary 1 holds: the invariant set N_{lambda^m x^n} is complete and irreducible for illumination and can be approximated by Eq. 5-6
    Proof deferred to supplementary material not included in v1; this supports the claim that N isolates light effects.
  • domain assumption Max-RGB illuminance map L = max_c I_c estimates illumination for k-means initialization
    Common practice used in Section 3.1 to obtain dark, well-lit, and highlight clusters.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Night-to-Day Translation via Illumination Degradation Disentanglement." pith.science (2026). https://pith.science/paper/NLY66TJV

@misc{pith2026241114504,
  author       = {Pith},
  title        = {Pith review of: Night-to-Day Translation via Illumination Degradation Disentanglement},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NLY66TJV}},
  note         = {Machine review of arXiv:2411.14504}
}
read the original abstract

Night-to-Day translation (Night2Day) aims to achieve day-like vision for nighttime scenes. However, processing night images with complex degradations remains a significant challenge under unpaired conditions. Previous methods that uniformly mitigate these degradations have proven inadequate in simultaneously restoring daytime domain information and preserving underlying semantics. In this paper, we propose \textbf{N2D3} (\textbf{N}ight-to-\textbf{D}ay via \textbf{D}egradation \textbf{D}isentanglement) to identify different degradation patterns in nighttime images. Specifically, our method comprises a degradation disentanglement module and a degradation-aware contrastive learning module. Firstly, we extract physical priors from a photometric model based on Kubelka-Munk theory. Then, guided by these physical priors, we design a disentanglement module to discriminate among different illumination degradation regions. Finally, we introduce the degradation-aware contrastive learning strategy to preserve semantic consistency across distinct degradation regions. Our method is evaluated on two public datasets, demonstrating a significant improvement in visual quality and considerable potential for benefiting downstream tasks.

Figures

Figures reproduced from arXiv: 2411.14504 by the authors.

Figure 1
Figure 1. Illustration of our motivation. (a) The disentanglement [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The first row displays nighttime images, while the second row shows the disentanglement results using 3-Cluster K-means. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. The overall architecture of the proposed N2D3 method. The training phase contains the physical prior informed degradation [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: The qualitative comparison results on the BDD100K dataset. [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: The qualitative comparison results on the Alderley dataset. [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 2
Figure 2. Figure 2: Visual comparison of the segmentation results with previous methods. he visual comparison of nighttime segmentation. The results demonstrate that our method learns a more ac proiding the greatest performance improement oer the tpical model trained on datime images and …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 25 canonical work pages

  1. [1]

    Night-to-day image translation for retrieval-based localization

    Asha Anoosheh, Torsten Sattler, Radu Timofte, Marc Polle- feys, and Luc Van Gool. Night-to-day image translation for retrieval-based localization. In 2019 International Confer- ence on Robotics and Automation (ICRA), pages 5958–5964. IEEE, 2019. 1, 2, 6, 7

    work page 2019

  2. [2]

    Rethinking atrous convolution for seman- tic image segmentation

    Liang-Chieh Chen, George Papandreou, Florian Schroff, and Hartwig Adam. Rethinking atrous convolution for seman- tic image segmentation. arXiv preprint arXiv:1706.05587 ,

    arXiv

  3. [3]

    Stargan: Unified genera- tive adversarial networks for multi-domain image-to-image translation

    Yunjey Choi, Minje Choi, Munyoung Kim, Jung-Woo Ha, Sunghun Kim, and Jaegul Choo. Stargan: Unified genera- tive adversarial networks for multi-domain image-to-image translation. In Proc. IEEE Conference on Computer Vision and Pattern Recognition, pages 8789–8797, 2018. 7

    work page 2018

  4. [4]

    Learning to see in nighttime driving scenes with inter- frequency priors

    Zhentao Fan, Xianhao Wu, Xiang Chen, and Yufeng Li. Learning to see in nighttime driving scenes with inter- frequency priors. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 4218–4225, 2023. 2

    work page 2023

  5. [5]

    J-M Geusebroek, Rein Van den Boomgaard, Arnold W. M. Smeulders, and Hugo Geerts. Color invariance. IEEE Trans- actions on Pattern analysis and machine intelligence , 23 (12):1338–1350, 2001. 3

    work page 2001

  6. [6]

    Zero-reference deep curve estimation for low-light image enhancement

    Chunle Guo, Chongyi Li, Jichang Guo, Chen Change Loy, Junhui Hou, Sam Kwong, and Runmin Cong. Zero-reference deep curve estimation for low-light image enhancement. In Proc. IEEE Conference on Computer Vision and Pattern Recognition, pages 1780–1789, 2020. 7

    work page 2020

  7. [7]

    Gans trained by a two time-scale update rule converge to a local nash equilib- rium

    Martin Heusel, Hubert Ramsauer, Thomas Unterthiner, Bernhard Nessler, and Sepp Hochreiter. Gans trained by a two time-scale update rule converge to a local nash equilib- rium. Advances in Neural Information Processing Systems , 30, 2017. 6

    work page 2017

  8. [8]

    Global structure-aware diffusion pro- cess for low-light image enhancement

    Jinhui Hou, Zhiyu Zhu, Junhui Hou, Hui Liu, Huanqiang Zeng, and Hui Yuan. Global structure-aware diffusion pro- cess for low-light image enhancement. Advances in Neural Information Processing Systems, 36, 2024. 7

    work page 2024

Show all 30 references
  1. [9]

    Qs-attn: Query-selected attention for contrastive learning in i2i translation

    Xueqi Hu, Xinyue Zhou, Qiusheng Huang, Zhengyi Shi, Li Sun, and Qingli Li. Qs-attn: Query-selected attention for contrastive learning in i2i translation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 18291–18300, 2022. 2

    2022

  2. [10]

    Memory-guided unsupervised image-to- image translation

    Somi Jeong, Youngjung Kim, Eungbean Lee, and Kwanghoon Sohn. Memory-guided unsupervised image-to- image translation. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 6558–6567, 2021. 2

    2021

  3. [11]

    Low-light image enhancement with wavelet-based diffusion models

    Hai Jiang, Ao Luo, Haoqiang Fan, Songchen Han, and Shuaicheng Liu. Low-light image enhancement with wavelet-based diffusion models. ACM Transactions on Graphics (TOG), 42(6):1–14, 2023. 7

    2023

  4. [12]

    Enlightengan: Deep light enhancement without paired supervision

    Yifan Jiang, Xinyu Gong, Ding Liu, Yu Cheng, Chen Fang, Xiaohui Shen, Jianchao Yang, Pan Zhou, and Zhangyang Wang. Enlightengan: Deep light enhancement without paired supervision. IEEE Transactions on Image Process- ing, 30:2340–2349, 2021. 7

    2021

  5. [13]

    2pcnet: Two-phase consistency training for day-to-night unsupervised domain adaptive object detec- tion

    Mikhail Kennerley, Jian-Gang Wang, Bharadwaj Veeravalli, and Robby T Tan. 2pcnet: Two-phase consistency training for day-to-night unsupervised domain adaptive object detec- tion. In Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition , pages 114...

  6. [14]

    U-gat-it: Unsupervised generative attentional networks with adaptive layer-instance normalization for image-to- image translation

    Junho Kim, Minjae Kim, Hyeonwoo Kang, and Kwanghee Lee. U-gat-it: Unsupervised generative attentional networks with adaptive layer-instance normalization for image-to- image translation. arXiv preprint arXiv:1907.10830, 2019. 7

    1907 arXiv

  7. [15]

    Instaformer: Instance-aware image-to-image translation with transformer

    Soohyun Kim, Jongbeom Baek, Jihye Park, Gyeongnyeon Kim, and Seungryong Kim. Instaformer: Instance-aware image-to-image translation with transformer. InProceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition, pages 18321–18331, 2022. 2

    2022

  8. [16]

    Adverse weather image translation with asymmetric and uncertainty-aware gan

    Jeong-gi Kwak, Youngsaeng Jin, Yuanming Li, Dongsik Yoon, Donghyeon Kim, and Hanseok Ko. Adverse weather image translation with asymmetric and uncertainty-aware gan. arXiv preprint arXiv:2112.04283, 2021. 1, 2, 7

    2021 arXiv

  9. [17]

    Milford and Gordon

    Michael J. Milford and Gordon. F. Wyeth. Seqslam: Visual route-based navigation for sunny summer days and stormy winter nights. In 2012 IEEE International Conference on Robotics and Automation, pages 1643–1649, 2012. 6

    2012

  10. [18]

    Contrastive learning for unpaired image-to-image translation

    Taesung Park, Alexei Efros, Richard Zhang, and Jun-Yan Zhu. Contrastive learning for unpaired image-to-image translation. In European Conference on Computer Vision , pages 319–345, 2020. 2, 6, 7

    2020

  11. [19]

    Shunit: Style harmonization for unpaired image-to-image translation

    Seokbeom Song, Suhyeon Lee, Hongje Seong, Kyoung- won Min, and Euntai Kim. Shunit: Style harmonization for unpaired image-to-image translation. Proceedings of the AAAI Conference on Artificial Intelligence, 37(2):2292– 2302, 2023. 2

    2023

  12. [20]

    Instance-wise hard negative example gen- eration for contrastive learning in unpaired image-to-image translation

    Weilun Wang, Wengang Zhou, Jianmin Bao, Dong Chen, and Houqiang Li. Instance-wise hard negative example gen- eration for contrastive learning in unpaired image-to-image translation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 14020–14029, 2021. 2

    2021

  13. [21]

    Stegogan: Leveraging steganography for non-bijective image-to-image translation

    Sidi Wu, Yizi Chen, Samuel Mermet, Lorenz Hurni, Konrad Schindler, Nicolas Gonthier, and Loic Landrieu. Stegogan: Leveraging steganography for non-bijective image-to-image translation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages...

    2024

  14. [22]

    Image- to-image translation for autonomous driving from coarsely- aligned image pairs

    Youya Xia, Josephine Monica, Wei-Lun Chao, Bharath Har- iharan, Kilian Q Weinberger, and Mark Campbell. Image- to-image translation for autonomous driving from coarsely- aligned image pairs. In 2023 IEEE International Confer- ence on Robotics and Automation (ICRA), pages 7756–...

    2023

  15. [23]

    Unsupervised image-to-image translation with density changing regular- ization

    Shaoan Xie, Qirong Ho, and Kun Zhang. Unsupervised image-to-image translation with density changing regular- ization. In Advances in Neural Information Processing Sys- tems, 2022. 2, 7

    2022

  16. [24]

    Unpaired image-to-image translation with shortest path reg- ularization

    Shaoan Xie, Yanwu Xu, Mingming Gong, and Kun Zhang. Unpaired image-to-image translation with shortest path reg- ularization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 10177–10187, 2023. 2, 7

    2023

  17. [25]

    Bdd100k: A diverse driving dataset for heterogeneous multitask learning

    Fisher Yu, Haofeng Chen, Xin Wang, Wenqi Xian, Yingying Chen, Fangchen Liu, Vashisht Madhavan, and Trevor Dar- rell. Bdd100k: A diverse driving dataset for heterogeneous multitask learning. In Proceedings of the IEEE/CVF con- ference on computer vision and pattern recognition ...

    2020

  18. [26]

    On the difficulty of unpaired infrared-to- visible video translation: Fine-grained content-rich patches transfer

    Zhenjie Yu, Shuang Li, Yirui Shen, Chi Harold Liu, and Shuigen Wang. On the difficulty of unpaired infrared-to- visible video translation: Fine-grained content-rich patches transfer. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) , ...

    2023

  19. [27]

    Modulated contrast for versatile image syn- thesis

    Fangneng Zhan, Jiahui Zhang, Yingchen Yu, Rongliang Wu, and Shijian Lu. Modulated contrast for versatile image syn- thesis. In Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition (CVPR), pages 18280– 18290, 2022. 2, 7

    2022

  20. [28]

    The unreasonable effectiveness of deep features as a perceptual metric

    Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shecht- man, and Oliver Wang. The unreasonable effectiveness of deep features as a perceptual metric. In Proceedings of the IEEE conference on computer vision and pattern recogni- tion, pages 586–595, 2018. 6

    2018

  21. [29]

    Fork- gan: Seeing into the rainy night

    Ziqiang Zheng, Yang Wu, Xinran Han, and Jianbo Shi. Fork- gan: Seeing into the rainy night. In European conference on computer vision, pages 155–170. Springer, 2020. 1, 2, 6, 7

    2020

  22. [30]

    Unpaired image-to-image translation using cycle-consistent adversarial networks

    Jun-Yan Zhu, Taesung Park, Phillip Isola, and Alexei Efros. Unpaired image-to-image translation using cycle-consistent adversarial networks. In Proc. IEEE International Confer- ence on Computer Vision, pages 2223–2232, 2017. 6, 7

    2017

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.