Pith. sign in

REVIEW 4 major objections 4 minor 79 references

Robust Adverse Weather Removal via Spectral-based Spatial Grouping

T0 review · 4 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read SSGformer removes rain, snow, and raindrops with one network by grouping image regions derived from Sobel edge and SVD low-frequency cues, reporting the best average PSNR on the All-Weather and WeatherStream benchmarks.

desk verdict A solid but incremental all-in-one weather removal architecture with a genuinely new spectral-plus-grouping design, marred by an under-specified grouping rule and small SOTA margins that need error bars. read the letter →

arxiv 2507.22498 v2 pith:LLAZB7RP submitted 2025-07-30 cs.CV cs.AI

Yuhwan Jeong , Yunseo Yang , Youngho Yoon , Kuk-Jin Yoon This is my paper
classification cs.CVcs.AI
keywords adverseweatherremovalall-in-oneimagerestorationspectraldecompositionSobeledgedetectionsingularvaluegroup-wiseattentionspatialgroupingtransformer
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 SSGformer, a single transformer network that restores images degraded by rain, snow, and raindrops. The central claim is that decomposing the degraded image into Sobel-detected edge features and SVD low-frequency components, fusing them, and using the fused features to partition the image into spatially similar groups makes a unified model handle diverse weather distortions better than global frequency filtering. On the All-Weather benchmark the model reports 32.63 dB average PSNR, the best overall result, and on the real-world WeatherStream benchmark it reports 24.77 dB average PSNR. The paper argues that the gain comes from the combination of the spectral prompt and the group-wise attention it enables, not from extra parameters.

What carries the argument

The load-bearing mechanism is the grouping-mask $M_p$, a single-channel map produced by a 7×7 convolution over the fused spectral feature $F_S$; it splits the feature map into $g_p$ groups of equal pixel count ordered by mask value. Feature-Grouped Attention projects each group into in-group and cross-group queries, keys, and a shared value, runs attention within a group and with the most similar other group chosen by cosine similarity, and combines the outputs with learnable weights. The spectral prompt makes the mask degradation-aware: Sobel captures high-frequency edges while SVD captures low-frequency structure, and linear attention lets the two cues inform each other before the mask is generated.

What would settle it

Replace the learned grouping mask with a fixed random partition of pixels into $g_p$ equal groups and retrain; if All-Weather PSNR stays near 32.63 dB, the mask is not the source of the gain. A cheaper check is to compute the same SSGformer's mask on an image and its horizontally flipped or brightness-shifted version and measure overlap: low overlap would indicate the groups are not stable degradation regions.

Watch

Extended reading notes

Core claim

SSGformer is an encoder-decoder transformer whose encoder is built from Spectral-based Decomposition Prompts, Mask Generators, and Spatial Grouping Transformer Blocks. The prompt runs the input through a Sobel operator and an SVD filter, refines both with separate blocks, fuses them with multi-head linear attention, and combines the results into a degradation-aware feature $F_S$. A single 7×7 convolution turns $F_S$ into a one-channel grouping mask $M_p$, which partitions the feature map at stage $p$ into $g_p$ equal-sized groups by mask value. Feature-Grouped Attention then applies in-group attention within each group and cross-group attention to the most similar group selected by cosine similarity over channel-pooled representatives, with learnable weights balancing the two streams; half the blocks attend along channels and half along space. The paper reports that this design reaches 32.63 dB PSNR on All-Weather and 24.77 dB on WeatherStream, with ablations showing that both the spectral prompt and the grouped attention contribute.

Load-bearing premise

The load-bearing premise is that the single-convolution grouping mask partitions pixels into groups that actually track similar degradation, so that running attention inside and across those groups is more useful than ordinary attention; if the mask assignments are arbitrary or unstable, the reported gains would vanish.

Editorial extensions

If this is right

  • On the All-Weather dataset, SSGformer reports 32.63 dB average PSNR and 0.939 average SSIM, the best overall scores, with per-task gains of +0.35 dB on rain and +0.18 dB on raindrop over the second-best method.
  • On the real-world WeatherStream dataset, SSGformer reports the best average PSNR at 24.77 dB, ahead of TransWeather, WGWS, and Histoformer.
  • Ablations show the full model beats variants with only Sobel or only SVD, only channel attention, and only in-group attention, so each proposed component contributes to the reported gain.
  • Because the model relies only on intra-image spectral and spatial cues, the approach is applicable without external knowledge sources such as LLMs or VLMs.

Reading between the lines

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

  • If the group selector's cosine-similarity choice is what drives the cross-group gain, then the same selector could be paired with any degradation-aware mask, including masks produced by learned prompt features; this would be a cheap test of whether the spectral prompt or the selection rule matters more.
  • The paper does not visualize the learned grouping masks or vary the group count $g_p$; measuring mask consistency across augmentations and reporting a group-count sensitivity curve would tell whether the mask is capturing degradation structure or just intensity thresholds.
  • Because Sobel and SVD are generic low-level cues, the same architecture could be pointed at other spatially varying corruptions such as motion blur, compression noise, or underwater scatter without retraining the spectral decomposer, though the paper only demonstrates weather.
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 / 4 minor

Summary. The paper proposes SSGformer, an all-in-one adverse weather removal network. It extracts high-frequency edge features via a Sobel operator and low-frequency features via singular value decomposition, fuses these with multi-head linear attention to produce a degradation-aware feature F_S, and generates a single-channel grouping mask from F_S using a convolution. The mask is then used to spatially group features and to feed a Spatial Grouping Transformer Block (SGTB) with in-group and cross-group attention, where the cross-group partner is selected by a cosine-similarity group selector. The model is evaluated on the All-weather and WeatherStream datasets, reporting the best average PSNR/SSIM on All-weather and WeatherStream, with the strongest margins on rain (+0.35 dB) and raindrop (+0.18 dB). The authors include ablations of the SDP components, attention configuration, group attention, and the combination of FGA and SDP, and they provide a public code link.

Significance. If the spatial grouping mechanism works as described, the paper makes a useful architectural contribution by connecting spectral prompts to spatially grouped attention in multi-weather restoration. The ablation study is reasonably structured and the code is promised, which aids reproducibility. However, the central novelty—the learned grouping mask and the group-wise attention driven by it—is not precisely specified, and the report does not demonstrate that the performance gains are attributable to the grouping itself rather than to the mask-as-gate or to added parameters. The reported margins over the second-best methods are small and are not accompanied by error bars or multiple-seed statistics, so the state-of-the-art claim is not yet rigorously supported.

major comments (4)
  1. [Sec. 3.4, Eq. (5) and Fig. 3(b)] The grouping operation is underspecified. The text states that features are 'divided evenly into gp groups' and that 'each grouped feature has the same spatial dimensions, i.e., the same number of elements,' but it does not define how the mask values are partitioned (e.g., sorting and equal-size binning), how ties are broken, or how cases where H*W is not divisible by gp are handled. Because the grouping is the core of the claimed contribution, a precise algorithmic definition is required.
  2. [Sec. 3.3 and Eq. (3)] The grouping assignment appears to be a hard, non-differentiable operation. If the mask values are sorted and split into equal-size bins, the assignment is piecewise constant and the gradient with respect to the mask values through the grouping step is zero almost everywhere. The only gradient path to the mask is the element-wise multiplication in Eq. (3), which trains the mask as a gating signal rather than as a clustering signal. The paper does not describe a straight-through estimator, a soft assignment, or a differentiable sorting mechanism, so it is not established that the mask is actually learned to group spatially similar degradation patterns. Please provide a differentiable formulation or an explicit training mechanism, and add an ablation that isolates the grouping assignment from the mask gating (e.g., compare against a fixed random grouping or a fixed spatial partition).
  3. [Tables 1 and 2] The reported improvements over the second-best method are small: +0.35 dB on rain, +0.18 dB on raindrop, and +0.13–0.16 dB on WeatherStream sub-tasks. No error bars, standard deviations, or multiple-seed results are reported, so the claim of state-of-the-art performance is not statistically substantiated. Please provide the mean and standard deviation over at least three independent training runs, or a statistical significance test.
  4. [Sec. 3 and implementation details] Several load-bearing hyperparameters are not specified: the number of groups gp per stage (only the example gp=4 is given), the SVD truncation rank used to compute the low-frequency component in the SDP, and the correlation loss weight β in Eq. (10). Without these values, the method is not reproducible, and the sensitivity of the results to gp and the SVD rank is unknown. Please report all hyperparameters and, ideally, a sensitivity analysis for the group count.
minor comments (4)
  1. [Abstract] The phrase 'To address these issue' should be 'To address these issues.'
  2. [Table 2] The header row of Table 2 contains a typo ('PNSR' instead of 'PSNR') and the table formatting appears corrupted; please correct it.
  3. [Sec. 3.2] 'Specially' should be 'Specifically.'
  4. [Supplementary] The text refers to the supplementary for the objective function details and WeatherStream visualizations, but the supplementary material is not included in this submission. Please ensure it is available to reviewers.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: SSGformer's state-of-the-art claim is supported by external benchmarks and ablations, not by a derivation that reduces to its own inputs.

full rationale

The paper's central claim is empirical: SSGformer achieves state-of-the-art performance on the All-Weather and WeatherStream datasets compared with external baselines (Tables 1 and 2). The novel grouping-mask is generated from spectral features via a learned convolution (Sec. 3.3), and its contribution is evaluated through ablations (Tables 3-6) against variants without it, rather than being defined in terms of the restoration output. The mask does not encode the target clean image or the final PSNR; it is a trained intermediate representation. No fitted parameter is renamed as a prediction, and no load-bearing result is justified solely by a self-citation. While the hard even partitioning of mask values may raise architectural or optimization concerns about gradient flow, that is a correctness risk, not circularity. The claimed performance gains are not forced by construction or by the paper's definitions, so the appropriate circularity score is 0.

Assumptions & free parameters 3 free parameters · 3 assumptions · 2 invented entities

The ledger lists hand-chosen hyperparameters and domain assumptions typical for an empirical deep-learning architecture paper. The grouping-mask and group selector are computational constructs whose value is demonstrated only within the paper. No physical constants or derivationally fitted coefficients are involved.

free parameters (3)
  • group count gp = 4 (illustrated)
    Number of spatial groups in Feature-Grouped Attention, chosen by hand and shown as 4 in Fig. 3(b); no sensitivity analysis or selection criterion is provided.
  • correlation loss weight beta = not reported
    Weight of the Pearson correlation loss in Eq. 10; tuned on validation but not specified in the main text or supplementary.
  • SVD truncation rank = not specified
    The SVD filter's rank truncation for low-frequency extraction is not described; it must be chosen to balance detail and smoothness, and it affects FSVD.
assumptions (3)
  • domain assumption Supervised paired training on synthetic datasets transfers to real-world degradation
    Standard in image restoration; the method is trained and evaluated on synthetic All-weather data, with qualitative examples on real images (Sec. 4.2).
  • domain assumption Sobel edge responses and SVD low-rank components capture degradation-relevant spectral information
    Core motivation for the Spectral-based Decomposition Prompt, supported only by in-paper ablations (Sec. 4.3), not by an independent theoretical or empirical argument.
  • domain assumption The Pearson correlation loss improves patch-level alignment and restoration quality
    Adopted from Histoformer [47] without independent justification in this paper; the total loss in Eq. 10 treats it as beneficial.
invented entities (2)
  • grouping-mask
    purpose: A single-channel map that partitions feature elements into groups for in-group and cross-group attention.
    Introduced in Sec. 3.3; its utility is evidenced only by internal ablations, with no external measure of mask quality or stability.
  • group selector
    purpose: Selects the most similar group via cosine similarity on pooled representative features, used to choose the query in cross-group attention.
    Introduced in Sec. 3.4 and Fig. 4; effectiveness shown only through Table 5 ablation, with no analysis of selection accuracy.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Robust Adverse Weather Removal via Spectral-based Spatial Grouping." pith.science (2026). https://pith.science/paper/LLAZB7RP

@misc{pith2026250722498,
  author       = {Pith},
  title        = {Pith review of: Robust Adverse Weather Removal via Spectral-based Spatial Grouping},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LLAZB7RP}},
  note         = {Machine review of arXiv:2507.22498}
}
read the original abstract

Adverse weather conditions cause diverse and complex degradation patterns, driving the development of All-in-One (AiO) models. However, recent AiO solutions still struggle to capture diverse degradations, since global filtering methods like direct operations on the frequency domain fail to handle highly variable and localized distortions. To address these issue, we propose Spectral-based Spatial Grouping Transformer (SSGformer), a novel approach that leverages spectral decomposition and group-wise attention for multi-weather image restoration. SSGformer decomposes images into high-frequency edge features using conventional edge detection and low-frequency information via Singular Value Decomposition. We utilize multi-head linear attention to effectively model the relationship between these features. The fused features are integrated with the input to generate a grouping-mask that clusters regions based on the spatial similarity and image texture. To fully leverage this mask, we introduce a group-wise attention mechanism, enabling robust adverse weather removal and ensuring consistent performance across diverse weather conditions. We also propose a Spatial Grouping Transformer Block that uses both channel attention and spatial attention, effectively balancing feature-wise relationships and spatial dependencies. Extensive experiments show the superiority of our approach, validating its effectiveness in handling the varied and intricate adverse weather degradations.

Figures

Figures reproduced from arXiv: 2507.22498 by the authors.

Figure 1
Figure 1. Overview. We propose SSGformer, an All-in-One ad￾verse weather removal model with three key components. (a) So￾bel operator and SVD extract spectral information from degraded image, producing degradation-aware features through spectral fea￾ture fusion. (b) The fused features identify degradation patterns and generate a “grouping-mask” for spatial grouping. (c) Features are grouped by the mask for group-wise spatial/… view at source ↗
Figure 2
Figure 2. Overall framework. (a) SSGformer is a transformer-based encoder-decoder network that restores a weather-degraded image ID to a clean image IC . The encoder consists of Spectral-based Decomposition Prompt (SDP), Mask Generator (MG), and Spatial Grouping Transformer Block (SGTB), which include two types of blocks: SGTB-C and SGTB-S. At stage p, there exist as many blocks as the number indicated by Lp. The stage p incr… view at source ↗
Figure 3
Figure 3. (a) Spatial Grouping Transformer Block (SGTB) starts with a convolution layer that fuses the previous stage fea￾tures Fp−1 and the degradation-aware features FS. The grouping￾mask Mp is element-wise multiplied with F ′ p and is also fed into Feature-Grouped Attention. (b) Illustration of Feature-Grouped Attention. Features are grouped using Mp, which encodes spatial similarity. In-group and cross-group attention are… view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Illustration of the group selector process. (1) Each group’s value is pooled to obtain a representative feature (R1∼4). (2) Cosine similarity is computed using the representative fea￾tures to measure inter-group similarity. The darker the matrix, the higher the similar…
Figure 5
Figure 5. Figure 5: Qualitative results of deraining on Outdoor-rain [ [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Qualitative results of raindrop removal on RainDrop [ [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Qualitative results of desnowing Snow100K-L [ [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: Real-world degradation removal examples. From [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

79 extracted references · 74 canonical work pages

  1. [1]

    Dehazenet: An end-to-end system for single image haze removal

    Bolun Cai, Xiangmin Xu, Kui Jia, Chunmei Qing, and Dacheng Tao. Dehazenet: An end-to-end system for single image haze removal. IEEE transactions on im- age processing, 25(11):5187–5198, 2016. 2

    work page 2016

  2. [2]

    Snow re- moval in video: A new dataset and a novel method

    Haoyu Chen, Jingjing Ren, Jinjin Gu, Hongtao Wu, Xuequan Lu, Haoming Cai, and Lei Zhu. Snow re- moval in video: A new dataset and a novel method. In 2023 IEEE/CVF International Conference on Com- puter Vision (ICCV) , pages 13165–13176. IEEE,

    work page 2023

  3. [3]

    Robust video content alignment and com- pensation for rain removal in a cnn framework

    Jie Chen, Cheen-Hau Tan, Junhui Hou, Lap-Pui Chau, and He Li. Robust video content alignment and com- pensation for rain removal in a cnn framework. InPro- ceedings of the IEEE conference on computer vision and pattern recognition, pages 6286–6295, 2018. 8

    work page 2018

  4. [4]

    Msp- former: Multi-scale projection transformer for single image desnowing

    Sixiang Chen, Tian Ye, Yun Liu, Taodong Liao, Jingxia Jiang, Erkang Chen, and Peng Chen. Msp- former: Multi-scale projection transformer for single image desnowing. In ICASSP 2023-2023 IEEE Inter- national Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 1–5. IEEE, 2023. 1, 2

    work page 2023

  5. [5]

    Jstasr: Joint size and transparency-aware snow removal algorithm based on modified partial convolution and veiling effect re- moval

    Wei-Ting Chen, Hao-Yu Fang, Jian-Jiun Ding, Cheng- Che Tsai, and Sy-Yen Kuo. Jstasr: Joint size and transparency-aware snow removal algorithm based on modified partial convolution and veiling effect re- moval. In Computer Vision–ECCV 2020: 16th Euro- pean Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXI 16 , pages 754–770. Springer,

    work page 2020

  6. [6]

    All snow removed: Single image desnowing al- gorithm using hierarchical dual-tree complex wavelet representation and contradict channel loss

    Wei-Ting Chen, Hao-Yu Fang, Cheng-Lin Hsieh, Cheng-Che Tsai, I Chen, Jian-Jiun Ding, Sy-Yen Kuo, et al. All snow removed: Single image desnowing al- gorithm using hierarchical dual-tree complex wavelet representation and contradict channel loss. In Pro- ceedings of the IEEE/CVF International Conference on Computer Vision, pages 4196–4205, 2021. 2

    work page 2021

  7. [7]

    Learning multiple adverse weather removal via two- stage knowledge learning and multi-contrastive regu- larization: Toward a unified model

    Wei-Ting Chen, Zhi-Kai Huang, Cheng-Che Tsai, Hao-Hsiang Yang, Jian-Jiun Ding, and Sy-Yen Kuo. Learning multiple adverse weather removal via two- stage knowledge learning and multi-contrastive regu- larization: Toward a unified model. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 17653–17662, 2022. 1, 3, 6

    work page 2022

  8. [8]

    Learning a sparse transformer network for effective image deraining

    Xiang Chen, Hao Li, Mingqiang Li, and Jinshan Pan. Learning a sparse transformer network for effective image deraining. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recog- nition, pages 5896–5905, 2023. 1, 2

    work page 2023

Show all 79 references
  1. [9]

    Hybrid cnn-transformer feature fusion for single image deraining

    Xiang Chen, Jinshan Pan, Jiyang Lu, Zhentao Fan, and Hao Li. Hybrid cnn-transformer feature fusion for single image deraining. In Proceedings of the AAAI conference on artificial intelligence , pages 378–386,

  2. [10]

    A generalized low- rank appearance model for spatio-temporally corre- lated rain streaks

    Yi-Lei Chen and Chiou-Ting Hsu. A generalized low- rank appearance model for spatio-temporally corre- lated rain streaks. In Proceedings of the IEEE inter- national conference on computer vision, pages 1968– 1975, 2013. 2

    1968

  3. [11]

    Pearson correlation coef- ficient

    Israel Cohen, Yiteng Huang, Jingdong Chen, Ja- cob Benesty, Jacob Benesty, Jingdong Chen, Yiteng Huang, and Israel Cohen. Pearson correlation coef- ficient. Noise reduction in speech processing , pages 1–4, 2009. 6

    2009

  4. [12]

    Selective frequency network for image restoration

    Yuning Cui, Yi Tao, Zhenshan Bing, Wenqi Ren, Xin- wei Gao, Xiaochun Cao, Kai Huang, and Alois Knoll. Selective frequency network for image restoration. In The Eleventh International Conference on Learning Representations, 2023. 1

    2023

  5. [13]

    Adair: Adaptive all-in-one image restoration via frequency mining and modulation

    Yuning Cui, Syed Waqas Zamir, Salman Khan, Alois Knoll, Mubarak Shah, and Fahad Shahbaz Khan. Adair: Adaptive all-in-one image restoration via frequency mining and modulation. arXiv preprint arXiv:2403.14614, 2024. 2, 4, 6

    2024 arXiv

  6. [14]

    An image is worth 16x16 words: Transformers for image recognition at scale

    Alexey Dosovitskiy. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929, 2020. 2

    2010 arXiv

  7. [15]

    Restor- ing an image taken through a window covered with dirt or rain

    David Eigen, Dilip Krishnan, and Rob Fergus. Restor- ing an image taken through a window covered with dirt or rain. In Proceedings of the IEEE international conference on computer vision, pages 633–640, 2013. 2

    2013

  8. [16]

    Removing rain from single images via a deep detail network

    Xueyang Fu, Jiabin Huang, Delu Zeng, Yue Huang, Xinghao Ding, and John Paisley. Removing rain from single images via a deep detail network. In Proceed- ings of the IEEE conference on computer vision and pattern recognition, pages 3855–3863, 2017. 2

    2017

  9. [17]

    Continual image deraining with hypergraph convolutional networks.IEEE Trans- actions on Pattern Analysis and Machine Intelligence, 45(8):9534–9551, 2023

    Xueyang Fu, Jie Xiao, Yurui Zhu, Aiping Liu, Feng Wu, and Zheng-Jun Zha. Continual image deraining with hypergraph convolutional networks.IEEE Trans- actions on Pattern Analysis and Machine Intelligence, 45(8):9534–9551, 2023. 2

    2023

  10. [18]

    Frequency- oriented efficient transformer for all-in-one weather- degraded image restoration

    Tao Gao, Yuanbo Wen, Kaihao Zhang, Jing Zhang, Ting Chen, Lidong Liu, and Wenhan Luo. Frequency- oriented efficient transformer for all-in-one weather- degraded image restoration. IEEE Transactions on Circuits and Systems for Video Technology , 34(3): 1886–1899, 2023. 1

    2023

  11. [19]

    Joint raindrop and haze removal from a single image

    Yina Guo, Jianguo Chen, Xiaowen Ren, Anhong Wang, and Wenwu Wang. Joint raindrop and haze removal from a single image. IEEE Transactions on Image Processing, 29:9508–9519, 2020. 2

    2020

  12. [20]

    Robust object de- tection in challenging weather conditions

    Himanshu Gupta, Oleksandr Kotlyar, Henrik An- dreasson, and Achim J Lilienthal. Robust object de- tection in challenging weather conditions. InProceed- ings of the IEEE/CVF Winter Conference on Applica- tions of Computer Vision, pages 7523–7532, 2024. 1

    2024

  13. [21]

    Image-to-image translation with conditional adversarial networks

    Phillip Isola, Jun-Yan Zhu, Tinghui Zhou, and Alexei A Efros. Image-to-image translation with conditional adversarial networks. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 1125–1134, 2017. 6

    2017

  14. [22]

    Au- tomatic single-image-based rain streaks removal via image decomposition

    Li-Wei Kang, Chia-Wen Lin, and Yu-Hsiang Fu. Au- tomatic single-image-based rain streaks removal via image decomposition. IEEE transactions on image processing, 21(4):1742–1755, 2011. 1, 2

    2011

  15. [23]

    Transformers are rnns: Fast autoregressive transformers with linear attention

    Angelos Katharopoulos, Apoorv Vyas, Nikolaos Pap- pas, and Franc ¸ois Fleuret. Transformers are rnns: Fast autoregressive transformers with linear attention. In International conference on machine learning , pages 5156–5165. PMLR, 2020. 4

    2020

  16. [24]

    Learning to remove bad weather: to- wards robust visual perception for self-driving

    Younkwan Lee, Yechan Kim, Jongmin Yu, and Moongu Jeon. Learning to remove bad weather: to- wards robust visual perception for self-driving. IEEE Robotics and Automation Letters, 2022. 1

    2022

  17. [25]

    Aod-net: All-in-one dehazing network

    Boyi Li, Xiulian Peng, Zhangyang Wang, Jizheng Xu, and Dan Feng. Aod-net: All-in-one dehazing network. In Proceedings of the IEEE international conference on computer vision, pages 4770–4778, 2017. 2

    2017

  18. [26]

    All-in-one image restora- tion for unknown corruption

    Boyun Li, Xiao Liu, Peng Hu, Zhongqin Wu, Jiancheng Lv, and Xi Peng. All-in-one image restora- tion for unknown corruption. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 17452–17462, 2022. 1, 3, 6

    2022

  19. [27]

    Heavy rain image restoration: Integrating physics model and conditional adversarial learning

    Ruoteng Li, Loong-Fah Cheong, and Robby T Tan. Heavy rain image restoration: Integrating physics model and conditional adversarial learning. In Pro- ceedings of the IEEE/CVF conference on computer vi- sion and pattern recognition, pages 1633–1642, 2019. 1, 2, 6, 7

    2019

  20. [28]

    All in one bad weather removal using architectural search

    Ruoteng Li, Robby T Tan, and Loong-Fah Cheong. All in one bad weather removal using architectural search. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 3175–3185, 2020. 2, 6

    2020

  21. [29]

    Prompt-in-prompt learn- ing for universal image restoration

    Zilong Li, Yiming Lei, Chenglong Ma, Junping Zhang, and Hongming Shan. Prompt-in-prompt learn- ing for universal image restoration. arXiv preprint arXiv:2312.05038, 2023. 4

    2023 arXiv

  22. [30]

    Im- proving image restoration through removing degrada- tions in textual representations

    Jingbo Lin, Zhilu Zhang, Yuxiang Wei, Dongwei Ren, Dongsheng Jiang, Qi Tian, and Wangmeng Zuo. Im- proving image restoration through removing degrada- tions in textual representations. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition, page...

    2024

  23. [31]

    Desnownet: Context-aware deep net- work for snow removal

    Yun-Fu Liu, Da-Wei Jaw, Shih-Chia Huang, and Jenq- Neng Hwang. Desnownet: Context-aware deep net- work for snow removal. IEEE Transactions on Image Processing, 27(6):3064–3073, 2018. 1, 2, 6, 8

    2018

  24. [32]

    Controlling vision- language models for universal image restoration

    Ziwei Luo, Fredrik K Gustafsson, Zheng Zhao, Jens Sj¨olund, and Thomas B Sch ¨on. Controlling vision- language models for universal image restoration. arXiv preprint arXiv:2310.01018, 3(8), 2023. 3

    2023 arXiv

  25. [33]

    Restoring vi- sion in adverse weather conditions with patch-based denoising diffusion models

    Ozan ¨Ozdenizci and Robert Legenstein. Restoring vi- sion in adverse weather conditions with patch-based denoising diffusion models. IEEE Transactions on Pattern Analysis and Machine Intelligence , 45(8): 10346–10357, 2023. 1

    2023

  26. [34]

    All-in-one image restoration for unknown degradations using adaptive discriminative filters for specific degradations

    Dongwon Park, Byung Hyun Lee, and Se Young Chun. All-in-one image restoration for unknown degradations using adaptive discriminative filters for specific degradations. In 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 5815–5824. IEEE, 2023

    2023

  27. [35]

    Multi- weather image restoration via domain translation

    Prashant W Patil, Sunil Gupta, Santu Rana, Svetha Venkatesh, and Subrahmanyam Murala. Multi- weather image restoration via domain translation. In Proceedings of the IEEE/CVF International Confer- ence on Computer Vision, pages 21696–21705, 2023

    2023

  28. [36]

    Promptir: Prompting for all-in-one image restoration

    Vaishnav Potlapalli, Syed Waqas Zamir, Salman H Khan, and Fahad Shahbaz Khan. Promptir: Prompting for all-in-one image restoration. Advances in Neural Information Processing Systems, 36, 2024. 1, 6

    2024

  29. [37]

    Attentive generative adversarial network for raindrop removal from a single image

    Rui Qian, Robby T Tan, Wenhan Yang, Jiajun Su, and Jiaying Liu. Attentive generative adversarial network for raindrop removal from a single image. InProceed- ings of the IEEE conference on computer vision and pattern recognition, pages 2482–2491, 2018. 1, 2, 6, 7

    2018

  30. [38]

    Enhanced pix2pix dehazing network

    Yanyun Qu, Yizi Chen, Jingying Huang, and Yuan Xie. Enhanced pix2pix dehazing network. InProceed- ings of the IEEE/CVF conference on computer vision and pattern recognition, pages 8160–8168, 2019. 1, 2

    2019

  31. [39]

    Re- moving raindrops and rain streaks in one go

    Ruijie Quan, Xin Yu, Yuanzhi Liang, and Yi Yang. Re- moving raindrops and rain streaks in one go. In Pro- ceedings of the IEEE/CVF conference on computer vi- sion and pattern recognition, pages 9147–9156, 2021. 8

    2021

  32. [40]

    Deep learning for seeing through window with rain- drops

    Yuhui Quan, Shijie Deng, Yixin Chen, and Hui Ji. Deep learning for seeing through window with rain- drops. In Proceedings of the IEEE/CVF International Conference on Computer Vision , pages 2463–2471,

  33. [41]

    Image desnowing via deep invertible separa- tion

    Yuhui Quan, Xiaoheng Tan, Yan Huang, Yong Xu, and Hui Ji. Image desnowing via deep invertible separa- tion. IEEE Transactions on Circuits and Systems for Video Technology, 33(7):3133–3144, 2023. 2

    2023

  34. [42]

    Single im- age dehazing via multi-scale convolutional neural net- works

    Wenqi Ren, Si Liu, Hua Zhang, Jinshan Pan, Xi- aochun Cao, and Ming-Hsuan Yang. Single im- age dehazing via multi-scale convolutional neural net- works. In Computer Vision–ECCV 2016: 14th Euro- pean Conference, Amsterdam, The Netherlands, Octo- ber 11-14, 2016, Proceedings, Pa...

    2016

  35. [43]

    Instant dehazing of images using po- larization

    Yoav Y Schechner, Srinivasa G Narasimhan, and Shree K Nayar. Instant dehazing of images using po- larization. In Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001, pages I–I. IEEE, 2001. 1

    2001

  36. [44]

    Uncertainty guided multi-scale attention network for raindrop removal from a single image

    Ming-Wen Shao, Le Li, De-Yu Meng, and Wang- Meng Zuo. Uncertainty guided multi-scale attention network for raindrop removal from a single image. IEEE Transactions on Image Processing , 30:4828– 4839, 2021. 2

    2021

  37. [45]

    Real-time single image and video super-resolution using an efficient sub-pixel convolu- tional neural network

    Wenzhe Shi, Jose Caballero, Ferenc Husz ´ar, Johannes Totz, Andrew P Aitken, Rob Bishop, Daniel Rueckert, and Zehan Wang. Real-time single image and video super-resolution using an efficient sub-pixel convolu- tional neural network. In Proceedings of the IEEE conference on com...

    2016

  38. [46]

    Vi- sion transformers for single image dehazing

    Yuda Song, Zhuqing He, Hui Qian, and Xin Du. Vi- sion transformers for single image dehazing. IEEE Transactions on Image Processing , 32:1927–1941,

    1927

  39. [47]

    Restoring images in adverse weather conditions via histogram transformer

    Shangquan Sun, Wenqi Ren, Xinwei Gao, Rui Wang, and Xiaochun Cao. Restoring images in adverse weather conditions via histogram transformer. In Eu- ropean Conference on Computer Vision , pages 111–

  40. [48]

    Maxim: Multi-axis mlp for image processing

    Zhengzhong Tu, Hossein Talebi, Han Zhang, Feng Yang, Peyman Milanfar, Alan Bovik, and Yinxiao Li. Maxim: Multi-axis mlp for image processing. In Pro- ceedings of the IEEE/CVF conference on computer vi- sion and pattern recognition, pages 5769–5780, 2022. 6

    2022

  41. [49]

    Transweather: Transformer-based restoration of images degraded by adverse weather conditions

    Jeya Maria Jose Valanarasu, Rajeev Yasarla, and Vishal M Patel. Transweather: Transformer-based restoration of images degraded by adverse weather conditions. In Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, pages 2353–2363, 2022. 1, 3, 6

    2022

  42. [50]

    Attention is all you need

    A Vaswani. Attention is all you need. Advances in Neural Information Processing Systems, 2017. 2

    2017

  43. [51]

    Spatial attentive single- image deraining with a high quality real rain dataset

    Tianyu Wang, Xin Yang, Ke Xu, Shaozhe Chen, Qiang Zhang, and Rynson WH Lau. Spatial attentive single- image deraining with a high quality real rain dataset. In Proceedings of the IEEE/CVF conference on com- puter vision and pattern recognition , pages 12270– 12279, 2019. 6

    2019

  44. [52]

    Smar- tassign: Learning a smart knowledge assignment strat- egy for deraining and desnowing

    Yinglong Wang, Chao Ma, and Jianzhuang Liu. Smar- tassign: Learning a smart knowledge assignment strat- egy for deraining and desnowing. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 3677–3686, 2023. 2

    2023

  45. [53]

    Contrastive learning for compact sin- gle image dehazing

    Haiyan Wu, Yanyun Qu, Shaohui Lin, Jian Zhou, Ruizhi Qiao, Zhizhong Zhang, Yuan Xie, and Lizhuang Ma. Contrastive learning for compact sin- gle image dehazing. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion (CVPR), pages 10551–10560, 2021. 2

    2021

  46. [54]

    Image de-raining transformer

    Jie Xiao, Xueyang Fu, Aiping Liu, Feng Wu, and Zheng-Jun Zha. Image de-raining transformer. IEEE Transactions on Pattern Analysis and Machine Intelli- gence, 45(11):12978–12995, 2022. 2

    2022

  47. [55]

    Raingan: Unsupervised raindrop removal via decomposition and composition

    Xu Yan and Yuan Ren Loke. Raingan: Unsupervised raindrop removal via decomposition and composition. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision , pages 14–23,

  48. [56]

    Language-driven all-in-one adverse weather removal

    Hao Yang, Liyuan Pan, Yan Yang, and Wei Liang. Language-driven all-in-one adverse weather removal. In Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition , pages 24902– 24912, 2024. 3, 8

    2024

  49. [57]

    Rain- drop removal with light field image using image in- painting

    Tao Yang, Xiaofei Chang, Hang Su, Nathan Crombez, Yassine Ruichek, Tomas Krajnik, and Zhi Yan. Rain- drop removal with light field image using image in- painting. IEEE Access, 8:58416–58426, 2020. 2

    2020

  50. [58]

    Semantic segmentation in multiple adverse weather conditions with domain knowledge retention

    Xin Yang, Wending Yan, Yuan Yuan, Michael Bi Mi, and Robby T Tan. Semantic segmentation in multiple adverse weather conditions with domain knowledge retention. In Proceedings of the AAAI Conference on Artificial Intelligence, pages 6558–6566, 2024. 1

    2024

  51. [59]

    Single image derain- ing using a recurrent multi-scale aggregation and en- hancement network

    Youzhao Yang and Hong Lu. Single image derain- ing using a recurrent multi-scale aggregation and en- hancement network. In 2019 IEEE International Con- ference on Multimedia and Expo (ICME), pages 1378–

    2019

  52. [60]

    Self-augmented un- paired image dehazing via density and depth decom- position

    Yang Yang, Chaoyue Wang, Risheng Liu, Lin Zhang, Xiaojie Guo, and Dacheng Tao. Self-augmented un- paired image dehazing via density and depth decom- position. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 2037–2046, 2022. 1, 2

    2022

  53. [61]

    Genuine knowl- edge from practice: Diffusion test-time adaptation for video adverse weather removal

    Yijun Yang, Hongtao Wu, Angelica I Aviles-Rivero, Yulun Zhang, Jing Qin, and Lei Zhu. Genuine knowl- edge from practice: Diffusion test-time adaptation for video adverse weather removal. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition, pa...

    2024

  54. [62]

    Uncertainty guided multi-scale residual learning-using a cycle spinning cnn for single image de-raining

    Rajeev Yasarla and Vishal M Patel. Uncertainty guided multi-scale residual learning-using a cycle spinning cnn for single image de-raining. In Proceed- ings of the IEEE/CVF conference on computer vision and pattern recognition, pages 8405–8414, 2019. 1, 2

    2019

  55. [63]

    Adverse weather removal with codebook pri- ors

    Tian Ye, Sixiang Chen, Jinbin Bai, Jun Shi, Cheng- hao Xue, Jingxia Jiang, Junjie Yin, Erkang Chen, and Yun Liu. Adverse weather removal with codebook pri- ors. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 12653–12664,

  56. [64]

    Rain streak removal based on non- negative matrix factorization

    Chia-Hung Yeh, Chih-Yang Lin, Kahlil Muchtar, and Pin-Hsian Liu. Rain streak removal based on non- negative matrix factorization. Multimedia Tools and Applications, 77:20001–20020, 2018. 1

    2018

  57. [65]

    Adherent raindrop modeling, detection and removal in video.IEEE trans- actions on pattern analysis and machine intelligence , 38(9):1721–1733, 2015

    Shaodi You, Robby T Tan, Rei Kawakami, Yasuhiro Mukaigawa, and Katsushi Ikeuchi. Adherent raindrop modeling, detection and removal in video.IEEE trans- actions on pattern analysis and machine intelligence , 38(9):1721–1733, 2015. 2

    2015

  58. [66]

    Multi-stage progressive image restoration

    Syed Waqas Zamir, Aditya Arora, Salman Khan, Munawar Hayat, Fahad Shahbaz Khan, Ming-Hsuan Yang, and Ling Shao. Multi-stage progressive image restoration. In Proceedings of the IEEE/CVF con- ference on computer vision and pattern recognition , pages 14821–14831, 2021. 6

    2021

  59. [67]

    Restormer: Efficient transformer for high- resolution image restoration

    Syed Waqas Zamir, Aditya Arora, Salman Khan, Mu- nawar Hayat, Fahad Shahbaz Khan, and Ming-Hsuan Yang. Restormer: Efficient transformer for high- resolution image restoration. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5728–5739...

    2022

  60. [68]

    Image de-raining using a conditional generative adver- sarial network

    He Zhang, Vishwanath Sindagi, and Vishal M Patel. Image de-raining using a conditional generative adver- sarial network. IEEE transactions on circuits and sys- tems for video technology , 30(11):3943–3956, 2019. 1, 2

    2019

  61. [69]

    Weatherstream: Light transport au- tomation of single image deweathering

    Howard Zhang, Yunhao Ba, Ethan Yang, Varan Mehra, Blake Gella, Akira Suzuki, Arnold Pfahnl, Chethan Chinder Chandrappa, Alex Wong, and Achuta Kadambi. Weatherstream: Light transport au- tomation of single image deweathering. In Proceed- ings of the IEEE/CVF Conference on Compu...

    2023

  62. [70]

    Ingredient- oriented multi-degradation learning for image restora- tion

    Jinghao Zhang, Jie Huang, Mingde Yao, Zizheng Yang, Hu Yu, Man Zhou, and Feng Zhao. Ingredient- oriented multi-degradation learning for image restora- tion. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 5825–5835, 2023. 1

    2023

  63. [71]

    Deep dense multi-scale network for snow removal using semantic and depth priors

    Kaihao Zhang, Rongqing Li, Yanjiang Yu, Wenhan Luo, and Changsheng Li. Deep dense multi-scale network for snow removal using semantic and depth priors. IEEE Transactions on Image Processing , 30: 7419–7431, 2021. 2, 6

    2021

  64. [72]

    Efficient deweahter mixture-of-experts with uncertainty-aware feature-wise linear modulation

    Rongyu Zhang, Yulin Luo, Jiaming Liu, Huan- rui Yang, Zhen Dong, Denis Gudovskiy, Tomoyuki Okuno, Yohei Nakata, Kurt Keutzer, Yuan Du, et al. Efficient deweahter mixture-of-experts with uncertainty-aware feature-wise linear modulation. In Proceedings of the AAAI Conference on ...

    2024

  65. [73]

    Fourmer: An efficient global modeling paradigm for image restoration

    Man Zhou, Jie Huang, Chun-Le Guo, and Chongyi Li. Fourmer: An efficient global modeling paradigm for image restoration. In International conference on machine learning, pages 42589–42601. PMLR, 2023. 2, 6

    2023

  66. [74]

    Unpaired image-to-image transla- tion using cycle-consistent adversarial networks

    Jun-Yan Zhu, Taesung Park, Phillip Isola, and Alexei A Efros. Unpaired image-to-image transla- tion using cycle-consistent adversarial networks. In Proceedings of the IEEE international conference on computer vision, pages 2223–2232, 2017. 6

    2017

  67. [75]

    Joint bi-layer optimization for single- image rain streak removal

    Lei Zhu, Chi-Wing Fu, Dani Lischinski, and Pheng- Ann Heng. Joint bi-layer optimization for single- image rain streak removal. InProceedings of the IEEE International Conference on Computer Vision (ICCV),

  68. [76]

    A fast single image haze removal algorithm using color at- tenuation prior

    Qingsong Zhu, Jiaming Mai, and Ling Shao. A fast single image haze removal algorithm using color at- tenuation prior. IEEE transactions on image process- ing, 24(11):3522–3533, 2015. 1

    2015

  69. [77]

    Mwformer: Multi-weather im- age restoration using degradation-aware transformers

    Ruoxi Zhu, Zhengzhong Tu, Jiaming Liu, Alan C Bovik, and Yibo Fan. Mwformer: Multi-weather im- age restoration using degradation-aware transformers. IEEE Transactions on Image Processing, 2024. 1

    2024

  70. [78]

    Learning weather-general and weather-specific fea- tures for image restoration under multiple adverse weather conditions

    Yurui Zhu, Tianyu Wang, Xueyang Fu, Xuanyu Yang, Xin Guo, Jifeng Dai, Yu Qiao, and Xiaowei Hu. Learning weather-general and weather-specific fea- tures for image restoration under multiple adverse weather conditions. In Proceedings of the IEEE/CVF conference on computer vision...

    2023

  71. [129]

    Springer, 2025. 3, 4, 6

    2025

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.