Pith. sign in

REVIEW 5 major objections 5 minor 46 references

WARLearn: Weather-Adaptive Representation Learning

T0 review · 5 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 WARLearn claims that a detector trained on clear weather can be adapted to fog and low light by fine-tuning only its feature backbone for 10 epochs on synthetic degraded images, aligning features with the Barlow Twins loss and adding zero…

desk verdict A zero-cost backbone adaptation that shows real-world gains on fog and low-light detection, but the stated mechanism is under-supported and the evaluation omits key baselines. read the letter →

arxiv 2411.14095 v1 pith:K2FHPNBC submitted 2024-11-21 cs.CV

Shubham Agarwal , Raz Birman , Ofer Hadar This is my paper · ORCID
classification cs.CV
keywords adverseweatherobjectdetectionrepresentationlearningBarlowTwinslossfeaturealignmentdomainadaptationfoggyimageslow-lightzero-parameter
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

WARLearn claims that a detector trained on clear weather can be ported to foggy and low-light conditions by fine-tuning only its feature-extraction backbone for 10 epochs on synthetic degraded copies of the same images, using Barlow Twins feature alignment against the frozen clean-weather backbone. The paper reports mean average precision (mAP@50) of 52.6% on the unseen real-world foggy RTTS dataset and 55.7% on the unseen real-world low-light ExDark dataset, surpassing dedicated enhancement-plus-detection pipelines. These gains come with zero additional parameters and zero added inference time, since only the backbone weights change. A sympathetic reader would care because the recipe offers a nearly free way to make existing clean-weather detectors usable in adverse conditions.

What carries the argument

The central machinery is a Barlow Twins cross-correlation loss computed between projected feature vectors from a frozen clean-weather reference backbone (the teacher) and a learnable adverse-weather backbone (the student), where the invariance term pushes diagonal correlations toward 1 and the redundancy-reduction term decorrelates off-diagonal elements. Paired with this are the synthetic degradation generators—an atmospheric scattering model for fog (Equations 5–8) and a gamma transformation for low light (Equation 9)—that create clean/degraded image pairs from PascalVOC for the alignment training.

What would settle it

Train WARLearn with the same recipe but change the synthetic degradation range, for example by generating fog with β values up to 0.3, and evaluate on a real dense-fog test set with visibility below 50 meters; if mAP@50 does not improve over the plain YOLOv3 baseline while simulated-fog mAP stays high, the representative-simulation assumption is falsified.

Watch

Extended reading notes

Core claim

On the paper's own terms, WARLearn establishes that aligning the backbone features of synthetic adverse-weather images to the fixed features of their clean counterparts with the Barlow Twins loss transfers to unseen real-world fog and darkness. The resulting hybrid model—the fine-tuned adverse-weather backbone combined with the original clean-weather prediction head—improves detection on RTTS and ExDark compared with YOLOv3 baselines and with prior frameworks, and does so without additional parameters or latency. The same procedure also improves YOLOv8, showing that the adaptation mechanism is not tied to one architecture.

Load-bearing premise

The synthetic weather models—atmospheric scattering with A=0.5 and β between 0.05 and 0.14 for fog, and gamma between 1.5 and 5 for low light—are representative enough of real-world fog and darkness that features aligned on synthetic data transfer to real images.

Editorial extensions

If this is right

  • A clean-weather detector can be ported to fog or low light with only 10 epochs of backbone fine-tuning, while keeping the original prediction head and suffering minimal accuracy loss on clear weather.
  • The framework adds no parameters and no inference time, making it compatible with real-time detection systems.
  • The same prediction head can be reused across weather conditions by swapping in the corresponding fine-tuned backbone, giving a modular deployment path.
  • The approach transfers across architectures, as demonstrated with both YOLOv3 and YOLOv8.
  • The redundancy-reduction term is essential: without it, mAP@50 on RTTS drops from 52.6% to 47.9%.

Reading between the lines

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

  • The same recipe likely extends to other degradations—rain, snow, blur, or sensor noise—because the framework only requires a simulator that produces paired clean and degraded versions of the same scene.
  • Because the teacher backbone is fixed, the student could potentially be aligned to unlabeled real target data through generative or style-transfer simulations, which might close part of the synthetic-to-real gap.
  • The flat mAP@50 across simulated fog levels suggests the fine-tuned backbone is robust within the training range, but real-world generalization may depend on how well that range covers the target distribution; testing on denser real fog would clarify this.
  • Combining WARLearn with enhancement preprocessors such as C2PNet and ZeroDCE gave intermediate results, indicating the two strategies are partially complementary rather than mutually exclusive.
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

5 major / 5 minor

Summary. WARLearn proposes a two-stage framework for adapting a clean-weather object detector to fog and low light. In the first stage a YOLOv3/YOLOv8 model is trained on clear Pascal VOC data. In the second stage a copy of the backbone (B_adv) is fine-tuned for 10 epochs on synthetic foggy or low-light versions of the same images, using a Barlow Twins loss between projected features of the fixed reference backbone (B_ref) and the adapted backbone; the projection head is discarded at inference. The final model combines B_adv with the original detection head. Experiments on simulated fog/low-light test sets and on the unseen real-world RTTS and ExDark datasets report mAP@50 improvements over several enhancement-based baselines (C2PNet, IA-YOLO, DENet, FeatEnhancer, PairLIE, ZeroDCE) with zero additional parameters and zero added inference time. Ablations include a YOLOv8 version, the effect of the redundancy-reduction term, and performance across fog levels.

Significance. If validated, WARLearn would provide a practical, low-cost way to port existing detectors to adverse weather without architectural modifications or extra inference compute, and the representation-learning formulation is clean. The paper's strengths include evaluation on unseen real-world datasets, a code release, and ablations that test the role of redundancy reduction and architecture generality. However, the empirical support is currently incomplete: the projector-based loss does not demonstrate raw-feature alignment, the closest domain-adaptation baselines are missing, and no variance estimates are reported. The real-world results are promising, but the margin over the nearest competitor is small, so the significance claim rests on statistical and methodological details that need to be addressed.

major comments (5)
  1. [Section 3, Eqs. (3)-(4)] The invariance loss is computed on projected features Z_ref and Z_adv, after the trainable projection network P_adv, which is discarded at inference. Minimizing (1 - Z_ref · Z_adv)^2 does not by itself establish that B_adv(x_adv) is close to B_ref(x_clean) in the raw feature space, because P_adv can learn a many-to-one map that aligns projected vectors without making the backbone features weather-invariant. To support the mechanistic claim in Section 3 that 'the weights of Badv learn to mitigate the effect of adverse conditions', the paper should provide raw-feature alignment statistics (e.g., cosine similarity between B_adv(x_adv) and B_ref(x_clean) before the projector) or an ablation with a fixed random projector.
  2. [Section 5.1 and Section 5.2, Tables 3-4] The comparison omits the most directly related domain-adaptation methods cited in the Introduction: DA-YOLO [43], Sindagi et al. [33], and MAET [2]. Because those methods also perform feature-level adaptation of detection backbones under adverse weather, the abstract's claim that WARLearn 'surpasses the performance of state-of-the-art frameworks' is not yet established. Add these baselines under the same evaluation protocol, or state explicitly why they are excluded.
  3. [Section 5, Tables 3-6, 9] No error bars, confidence intervals, or multiple-seed runs are reported. The real-world margins are modest (52.60 vs. 50.50 on RTTS; 55.70 vs. 53.70 on ExDark; 60.00 vs. 57.70 for YOLOv8 low-light), and without variance estimates the improvements cannot be distinguished from noise. Report at least three independent runs with mean and standard deviation, or an appropriate statistical test.
  4. [Section 4 and Section 5] The SimFoggy and SimLL test sets are generated with the same atmospheric-scattering and gamma-correction models (Eqs. (5)-(9)) used to create the fine-tuning data. Consequently the large simulated-test gains (75.10 and 70.90) partly reflect training-distribution overlap, not independent evidence of robustness. The unseen real-world RTTS and ExDark results are the load-bearing evidence for transfer; the synthetic results should be reframed as a sanity check or paired with a different simulation protocol.
  5. [Section 4, paragraph beginning 'Many of the SOTA methods...'] The claim that other methods were retrained on the same data as WARLearn is not backed by a documented protocol. Without specifying learning rate, epochs, optimizer, and any hyperparameter search for each baseline, an undocumented retraining procedure may disadvantage the baselines. Provide the exact configuration, or cite the original numbers with a clear statement of what was retrained.
minor comments (5)
  1. [Abstract] The phrase 'in-variance principal' should be 'invariance principle'.
  2. [Figure 4 and Section 5.2] The caption and text refer to 'sample images from the W ARLearn dataset'; this should be the ExDark dataset.
  3. [Section 5.2] The sentence 'do not generalize well on unseen real world foggy RTTS dataset' appears in the low-light results section and should refer to the ExDark dataset.
  4. [Table 7 caption and Table 8 caption] There are typos: 'parameteres' should be 'parameters' and 'Performace' should be 'Performance'.
  5. [Section 6.2] The sentence 'The results of this analysis underscore are provided in Table 7' is grammatically incorrect; remove 'underscore'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: headline results are on unseen real-world benchmarks, and the same-simulation synthetic tests are held-out evaluations, not fitted predictions.

full rationale

The paper's central claims are real-world mAP numbers on RTTS and ExDark, which are explicitly unseen and not used in training or fine-tuning, so the headline performance is independent evidence for the method. The synthetic SimFoggy and SimLL evaluations are generated from the same atmospheric-scattering and gamma-correction models used to create the training data, but they use held-out PascalVOC test images; this is a standard same-distribution evaluation and not a fitted-parameter result that is forced by construction. The training objective is defined on projected features through the trainable P_adv, and the paper infers that B_adv itself becomes weather-invariant; this is an empirical mechanism claim that could be challenged experimentally, but it is not a circular reduction because the loss is not defined as the equality of raw B_adv and B_ref features. The load-bearing references, including Barlow Twins and the weather simulation equations, are external prior works rather than self-citations by the present authors, and no uniqueness theorem or ansatz is imported from the authors' own prior work. Therefore, no equation-level or definition-level circularity is exhibited, and the paper is self-contained against external benchmarks.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The method introduces no new physical entities. It relies on standard math (Barlow loss) and two domain assumptions about synthetic weather simulation. The main hand-set choices are the loss weight, simulation ranges, and fine-tuning schedule; the paper provides a single ablation for lambda but no sensitivity analysis for the simulation parameters or training schedule.

free parameters (4)
  • lambda (Barlow loss weight) = 0.001
    Chosen from the original Barlow Twins paper [42]; the ablation (Table 9) shows the redundancy reduction term has a large effect, so the central results depend on this hand-set value.
  • fog simulation beta range = 0.05 to 0.14 (beta = 0.01*i + 0.05, i=0..9)
    Hand-chosen range for synthetic fog generation; the realism of this simulation is a load-bearing assumption for transfer to RTTS.
  • low-light gamma range = uniform [1.5, 5]
    Hand-chosen range for synthetic low-light generation; the realism of this simulation is load-bearing for transfer to ExDark.
  • fine-tuning epochs and learning rate = 10 epochs, LR 0.0002 (YOLOv3); 10 epochs, LR 0.0001 (YOLOv8)
    Hand-set hyperparameters; no sensitivity analysis is provided.
assumptions (5)
  • domain assumption The atmospheric scattering model (Eq. 5-7) with A=0.5 and beta in [0.05,0.14] produces fog that is representative of real-world fog in RTTS.
    Used to generate all synthetic fog training data (Section 4); if this simulation mismatches real fog, the fine-tuned features will not transfer.
  • domain assumption Gamma correction (Eq. 9) with gamma in [1.5,5] produces low-light images representative of ExDark.
    Used to generate all synthetic low-light training data (Section 4).
  • domain assumption Aligning adverse-weather features to clear-weather features via the Barlow loss is sufficient to improve detection with the original prediction head.
    Core premise of the framework (Section 3); if the original head is incompatible with aligned features, the hybrid model would fail.
  • domain assumption The feature extractor and prediction head are separable, and fine-tuning only the backbone does not require re-tuning the head.
    Stated as a condition in Section 3; the method combines Badv with Pref without joint fine-tuning.
  • standard math Barlow Twins loss definitions (Eq. 1-2) are standard.
    Taken from [42]; no new mathematical content.

how reviews work

0 comments
Cite this review

Pith. "Pith review of WARLearn: Weather-Adaptive Representation Learning." pith.science (2026). https://pith.science/paper/K2FHPNBC

@misc{pith2026241114095,
  author       = {Pith},
  title        = {Pith review of: WARLearn: Weather-Adaptive Representation Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/K2FHPNBC}},
  note         = {Machine review of arXiv:2411.14095}
}
read the original abstract

This paper introduces WARLearn, a novel framework designed for adaptive representation learning in challenging and adversarial weather conditions. Leveraging the in-variance principal used in Barlow Twins, we demonstrate the capability to port the existing models initially trained on clear weather data to effectively handle adverse weather conditions. With minimal additional training, our method exhibits remarkable performance gains in scenarios characterized by fog and low-light conditions. This adaptive framework extends its applicability beyond adverse weather settings, offering a versatile solution for domains exhibiting variations in data distributions. Furthermore, WARLearn is invaluable in scenarios where data distributions undergo significant shifts over time, enabling models to remain updated and accurate. Our experimental findings reveal a remarkable performance, with a mean average precision (mAP) of 52.6% on unseen real-world foggy dataset (RTTS). Similarly, in low light conditions, our framework achieves a mAP of 55.7% on unseen real-world low light dataset (ExDark). Notably, WARLearn surpasses the performance of state-of-the-art frameworks including FeatEnHancer, Image Adaptive YOLO, DENet, C2PNet, PairLIE and ZeroDCE, by a substantial margin in adverse weather, improving the baseline performance in both foggy and low light conditions. The WARLearn code is available at https://github.com/ShubhamAgarwal12/WARLearn

Figures

Figures reproduced from arXiv: 2411.14095 by the authors.

Figure 1
Figure 1. Our proposed WARLearn framework initiates with the [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. In the Barlow Twins approach, we input two distorted [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Comparison of detection results between YOLOv3 base [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Comparison of detection results between YOLOv3 base [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Detection results with WARLearn framework on an image with different levels of foggy degradation indicated by [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

46 extracted references · 43 canonical work pages

  1. [43]

    Domain adaptive yolo for one-stage cross-domain detection

    Shizhao Zhang, Hongya Tuo, Jian Hu, and Zhongliang Jing. Domain adaptive yolo for one-stage cross-domain detection. In Asian Conference on Machine Learning, pages 785–797. PMLR, 2021. 1

    work page 2021

  2. [33]

    Prior-based domain adaptive object detec- tion for hazy and rainy conditions

    Vishwanath A Sindagi, Poojan Oza, Rajeev Yasarla, and Vishal M Patel. Prior-based domain adaptive object detec- tion for hazy and rainy conditions. InECCV, pages 763–780. Springer, 2020. 1, 2

    work page 2020

  3. [2]

    Multitask aet with orthogonal tangent regularity for dark object detection

    Ziteng Cui, Guo-Jun Qi, Lin Gu, Shaodi You, Zenghui Zhang, and Tatsuya Harada. Multitask aet with orthogonal tangent regularity for dark object detection. In Proceedings of the IEEE/CVF International Conference on Computer Vi- sion (ICCV), pages 2553–2562, October 2021. 1

    work page 2021

  4. [1]

    A. Cantor. Optics of the atmosphere–scattering by molecules and particles. IEEE Journal of Quantum Electronics , 14(9):698–699, 1978. 4

    work page 1978

  5. [3]

    Unsupervised domain adaptation for video transform- ers in action recognition

    Victor G Turrisi Da Costa, Giacomo Zara, Paolo Rota, Thi- ago Oliveira-Santos, Nicu Sebe, Vittorio Murino, and Elisa Ricci. Unsupervised domain adaptation for video transform- ers in action recognition. In 2022 26th International Con- ference on Pattern Recognition (ICPR) , pages 1258–1265. IEEE, 2022. 3

    work page 2022

  6. [4]

    Cf-yolo: Cross fusion yolo for object detection in ad- verse weather with a high-quality real snow dataset

    Qiqi Ding, Peng Li, Xuefeng Yan, Ding Shi, Luming Liang, Weiming Wang, Haoran Xie, Jonathan Li, and Mingqiang Wei. Cf-yolo: Cross fusion yolo for object detection in ad- verse weather with a high-quality real snow dataset. IEEE Transactions on Intelligent Transportation Systems, 2023. 1

    work page 2023

  7. [5]

    Multi-scale boosted de- hazing network with dense feature fusion

    Hang Dong, Jinshan Pan, Lei Xiang, Zhe Hu, Xinyi Zhang, Fei Wang, and Ming-Hsuan Yang. Multi-scale boosted de- hazing network with dense feature fusion. In CVPR, pages 2154–2164, 2020. 2

    work page 2020

  8. [6]

    Everingham., L

    M. Everingham., L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman. The PASCAL Visual Object Classes Challenge 2007 (VOC2007) Results. http://www.pascal- network.org/challenges/VOC/voc2007/workshop/index.html. 4

    work page 2007

Show all 46 references
  1. [7]

    Everingham., L

    M. Everingham., L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman. The PASCAL Visual Object Classes Challenge 2012 (VOC2012) Results. http://www.pascal- network.org/challenges/VOC/voc2012/workshop/index.html. 4

    2012

  2. [8]

    Learning a simple low-light image enhancer from paired low-light instances

    Zhenqi Fu, Yan Yang, Xiaotong Tu, Yue Huang, Xinghao Ding, and Kai-Kuang Ma. Learning a simple low-light image enhancer from paired low-light instances. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 22252–22261, 2023. 1, 6, 7

    2023

  3. [9]

    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. 2021. 2

    2021

  4. [10]

    Zero- reference deep curve estimation for low-light image en- hancement

    Chunle Guo Guo, Chongyi Li, Jichang Guo, Chen Change Loy, Junhui Hou, Sam Kwong, and Runmin Cong. Zero- reference deep curve estimation for low-light image en- hancement. In Proceedings of the IEEE conference on com- puter vision and pattern recognition (CVPR) , pages 1780– 17...

    2020

  5. [11]

    Fog simulation on real lidar point clouds for 3d object detection in adverse weather

    Martin Hahner, Christos Sakaridis, Dengxin Dai, and Luc Van Gool. Fog simulation on real lidar point clouds for 3d object detection in adverse weather. pages 15283–15292,

  6. [12]

    Featenhancer: En- hancing hierarchical features for object detection and beyond under low-light vision

    Khurram Azeem Hashmi, Goutham Kallempudi, Didier Stricker, and Muhammad Zeshan Afzal. Featenhancer: En- hancing hierarchical features for object detection and beyond under low-light vision. In Proceedings of the IEEE/CVF In- ternational Conference on Computer Vision (ICCV) , p...

    2023

  7. [13]

    Kenk, Khan Muham- mad, and Shervin Minaee

    Mahmoud Hassaballah, Mourad A. Kenk, Khan Muham- mad, and Shervin Minaee. Vehicle detection and tracking in adverse weather using a deep learning framework. In IEEE transactions on intelligent transportation systems 22 , pages 4230–4242, 2020. 1, 2

    2020

  8. [14]

    Cnn-based lidar point cloud de-noising in adverse weather

    Robin Heinzler, Florian Piewak, Philipp Schindler, and Wil- helm Stork. Cnn-based lidar point cloud de-noising in adverse weather. IEEE Robotics and Automation Letters , 10.1109/LRA.2020.2972865:2377–3774, 2020. 2

    2020

  9. [15]

    Dsnet: Joint semantic learning for object detection in inclement weather conditions

    Shih-Chia Huang, Trung-Hieu Le, , and Da-Wei Jaw. Dsnet: Joint semantic learning for object detection in inclement weather conditions. IEEE Transactions on Pattern Analysis and Machine Intelligence, 43(8):2623–2633, 2020. 2

    2020

  10. [16]

    Da-raw: Domain adaptive object detection for real-world adverse weather con- ditions

    Minsik Jeon, Junwon Seo, and Jihong Min. Da-raw: Domain adaptive object detection for real-world adverse weather con- ditions. ArXiv, abs/2309.08152, 2023. 1

    2023 arXiv

  11. [17]

    Ultralytics YOLO., Jan

    Glenn Jocher, Ayush Chaurasia, and Jing Qiu. Ultralytics YOLO., Jan. 2023. 2, 5, 7

    2023

  12. [18]

    Madhava Krishna

    Sanket Kalwar, Dhruv Patel, Aakash Aanegola, Kr- ishna Reddy Konda, Sourav Garg, and K. Madhava Krishna. Gdip: Gated differentiable image processing for object de- tection in adverse conditions. In ICRA, pages 7083–7089,

  13. [19]

    Kingma and Jimmy Lei Ba

    Diederik P. Kingma and Jimmy Lei Ba. Adam: A mthod for stochastic optimization. 2015. 4

    2015

  14. [20]

    Benchmarking single- image dehazing and beyond

    Boyi Li, Wenqi Ren, Dengpan Fu, Dacheng Tao, Dan Feng, Wenjun Zeng, and Zhangyang Wang. Benchmarking single- image dehazing and beyond. IEEE Transactions on Image Processing, 28(1):492–505, 2019. 4, 5

    2019

  15. [21]

    Detection- friendly dehazing: Object detection in real-world hazy scenes

    Chengyang Li, Heng Zhou, Yang Liu, Caidong Yang, Yongqiang Xie, Zhongbo Li, and Liping Zhu. Detection- friendly dehazing: Object detection in real-world hazy scenes. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023. 1

    2023

  16. [22]

    Domain adaptive object detection for au- tonomous driving under foggy weather

    Jinlong Li, Runsheng Xu, Jin Ma, Qin Zou, Jiaqi Ma, and Hongkai Yu. Domain adaptive object detection for au- tonomous driving under foggy weather. In Proceedings of the IEEE/CVF Winter Conference on Applications of Com- puter Vision, pages 612–622, 2023. 1

    2023

  17. [23]

    Image-adaptive yolo for object detec- tion in adverse weather conditions

    Wenyu Liu, Gaofeng Ren, Runsheng Yu, Shi Guo, Jianke Zhu, and Lei Zhang. Image-adaptive yolo for object detec- tion in adverse weather conditions. In Proceedings of the AAAI Conference on Artificial Intelligence, 2022. 1, 2, 5, 7

    2022

  18. [24]

    Getting to know low- light images with the exclusively dark dataset

    Yuen Peng Loh and Chee Seng Chan. Getting to know low- light images with the exclusively dark dataset. Computer Vision and Image Understanding, 178:30–42, 2019. 4, 6

    2019

  19. [25]

    Ecker, Matthias Bethge, and Wieland Brendel

    Claudio Michaelis, Benjamin Mitzkus, Robert Geirhos, Evgenia Rusak, Oliver Bringmann, Alexander S. Ecker, Matthias Bethge, and Wieland Brendel. Benchmarking ro- bustness in object detection: Autonomous driving when win- ter is coming. arXiv preprint arXiv:1907.07484, 2019. 2

    1907 arXiv

  20. [26]

    Narasimhan and S.K

    S.G. Narasimhan and S.K. Nayar. Vision and the at- mosphere. International Journal of Computer Vision , 14(9):233–254, 2002. 4

    2002

  21. [27]

    Narasimhan and S.K

    S.G. Narasimhan and S.K. Nayer. Contrast restoration of weather degraded images. IEEE Transactions on Pattern Analysis and Machine Intelligence, 25(6):713–724, 2003. 2

    2003

  22. [28]

    Black box few-shot adaptation for vision- language models, 2023

    Yassine Ouali, Adrian Bulat, Brais Martinez, and Georgios Tzimiropoulos. Black box few-shot adaptation for vision- language models, 2023. 3

    2023

  23. [29]

    Denet: Detection-driven enhancement network for object detection under adverse weather conditions

    Qingpao Qin, Kan Chang, Mengyuan Huang, and Guiqing Li. Denet: Detection-driven enhancement network for object detection under adverse weather conditions. 2022. 1, 2, 5, 7

    2022

  24. [30]

    Prior-guided source-free domain adaptation for human pose estimation

    Dripta S Raychaudhuri, Calvin-Khang Ta, Arindam Dutta, Rohit Lal, and Amit K Roy-Chowdhury. Prior-guided source-free domain adaptation for human pose estimation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 14996–15006, 2023. 3

    2023

  25. [31]

    Yolov3: An incremental improvement

    Joseph Redmon and Ali Farhadi. Yolov3: An incremental improvement. arXiv, 2018. 2, 4

    2018

  26. [32]

    Object de- tection in foggy weather conditions

    Prithwish Sen, Anindita Das, and Nilkanta Sahu. Object de- tection in foggy weather conditions. InInternational Confer- ence on Intelligent Computing & Optimization , pages 728– 737, 2021. 2

    2021

  27. [34]

    Robby T. Tan. Visibility in bad weather from a single image. In CVPR, pages 1–8, 2008. 1, 2

    2008

  28. [35]

    Exploring the application of large- scale pre-trained models on adverse weather removal

    Zhentao Tan, Yue Wu, Qiankun Liu, Qi Chu, Le Lu, Jieping Ye, and Nenghai Yu. Exploring the application of large- scale pre-trained models on adverse weather removal. arXiv preprint arXiv:2306.09008, 2019. 2

    2019 arXiv

  29. [36]

    Jeya Maria Jose Valanarasu, Rajeev Yasarla, and Vishal M. Patel. Transweather: Transformer-based restoration of im- ages degraded by adverse weather conditions. 2022. 2

    2022

  30. [37]

    R-yolo: A robust object detector in ad- verse weather

    Lucai Wang, Qin Hongda, Zhou Xuanyu, Lu Xiao, , and Zhang Fengting. R-yolo: A robust object detector in ad- verse weather. In IEEE Transactions on Instrumentation and Measurement 72, pages 1–11, 2022. 1, 2, 4

    2022

  31. [38]

    Deep retinex decomposition for low-light enhancement

    Chen Wei, Wenjing Wang, Wenhan Yang, and Jiaying Liu. Deep retinex decomposition for low-light enhancement. arXiv preprint arXiv:1808.04560, 2018. 2

    2018 arXiv

  32. [39]

    Contrastive learning-based robust ob- ject detection under smoky conditions

    Wei Wu, Hao Chang, Yonghua Zheng, Zhu Li, Zhiwen Chen, and Ziheng Zhang. Contrastive learning-based robust ob- ject detection under smoky conditions. In CVPR Workshops, pages 4295–4302, 2022. 2

    2022

  33. [40]

    Adverse weather target detection algorithm based on adaptive color levels and improved yolov5

    Jiale Yao, Fan Xiangsuo, Li Bing, and Qin Wenlin. Adverse weather target detection algorithm based on adaptive color levels and improved yolov5. InSensors 22, volume 21, 2022. 1, 2

    2022

  34. [41]

    Ad- verse weather removal with codebook priors

    Tian Ye, Sixiang Chen, Jinbin Bai, Jun Shi, Chenghao Xue, Jingxia Jiang, Junjie Yin, Erkang Chen, and Yun Liu. Ad- verse weather removal with codebook priors. 2023. 2

    2023

  35. [42]

    Barlow twins: Self-supervised learning via redundancy reduction

    Jure Zbontar, Li Jing, Ishan Misra, Yann LeCun, and St´ephane Deny. Barlow twins: Self-supervised learning via redundancy reduction. In ICML, 2021. 2, 4

    2021

  36. [44]

    Kindling the darkness: A practical low-light image enhancer

    Yonghua Zhang, Jiawan Zhang, and Xiaojie Guo. Kindling the darkness: A practical low-light image enhancer. In Pro- ceedings of the 27th ACM international conference on mul- timedia, pages 1632–1640, 2019. 2

    2019

  37. [45]

    Curricular contrastive regularization for physics- aware single image dehazing

    Yu Zheng, Jiahui Zhan, Shengfeng He, Junyu Dong, and Yong Du. Curricular contrastive regularization for physics- aware single image dehazing. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023. 1, 2, 5, 7

    2023

  38. [46]

    Represent, compare, and learn: A similarity-aware framework for class- agnostic counting

    Yurui Zhu, Tianyu Wang, Xueyang Fu, Xuanyu Yang, Xin Guo, Jifeng Dai, Yu Qiao, and Xiaowei Hu. Represent, compare, and learn: A similarity-aware framework for class- agnostic counting. In CVPR, 2023. 2

    2023

Pith tools

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