Pith. sign in

REVIEW 4 major objections 5 minor 37 references

Is it Raining Outside? Detection of Rainfall using General-Purpose Surveillance Cameras

T0 review · 4 major / 5 minor · reviewed 2026-08-14 · 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 a fully-convolutional 3D CNN detects rainfall from surveillance cameras only on scenes resembling its training data, failing entirely on an asphalt surface.

desk verdict Genuine dataset and baseline, but the abstract overclaims on the paper's own numbers. read the letter →

arxiv 1908.04034 v2 pith:5EHMMII7 submitted 2019-08-12 cs.CV

Joakim Bruslund Haurum , Chris H. Bahnsen , Thomas B. Moeslund This is my paper · ORCID
classification cs.CV
keywords raindetectionsurveillancecameras3DCNNC3DrainfallmeasurementVIRADAdatasetcomputervisionfullyconvolutionalnetwork
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

The paper asks whether ordinary surveillance cameras, not weather-beaten special-purpose setups, can serve as rain detectors. To answer it, the authors introduce the AAU Visual Rain Dataset (VIRADA), 215 hours of video from two Danish traffic crossings with rainfall labels from a laser disdrometer. They reimplement the previous state-of-the-art, Bossu et al.'s histogram-of-orientation-of-streaks algorithm, and compare it with a fully-convolutional 3D CNN (C3D) adapted from action recognition. On the crossing used for training, the CNN clearly outperforms Bossu's method, and it retains that edge on a similarly textured region of another crossing. On a differently textured asphalt surface, however, the CNN predicts no rain at all, showing that texture generalization, not rain detection per se, is the unresolved obstacle.

What carries the argument

The central object is the modified C3D network: a 3D convolutional architecture that takes 16-frame clips as input and ends in two 2D convolutional layers plus global averaging, converting it into a fully-convolutional network that scores every 112×112 crop of the frame. It is trained from scratch as a binary rain classifier. The comparison baseline is Bossu's pipeline, which extracts candidate rain streaks via the photometric constraint, builds a histogram of streak orientations, fits a Gaussian-uniform mixture with expectation-maximization, and uses a Kalman filter to smooth detections. The C3D's temporal receptive field is what lets it in principle see rain streaks moving across frames, while the Bossu method explicitly models streak orientation.

What would settle it

If a disdrometer were placed directly beside one of the cameras, and the current model's predictions disagreed with that local ground truth on a large fraction of rain minutes, the assumption tying distant gauge data to the camera scene would be falsified. Alternatively, manually labeling rain visibility on Crossing2-asphalt and showing that the C3D-FCN predicts no rain for clips with clearly visible streaks would falsify the claim that the network detects rain in that setting.

Watch

Extended reading notes

Core claim

On its own terms, the paper establishes that a 3D CNN trained on 16-frame clips of one traffic crossing can detect rainfall in surveillance footage from that crossing with high accuracy (MCC 0.94 on training, 0.58 on validation), comfortably beating the hand-crafted Bossu detector whose MCC stays near zero. The same network, applied as a fully-convolutional network with global averaging, also outperforms Bossu on a brick-textured region of a second crossing but fails completely on an asphalt region of that crossing, where it outputs 'no rain' for every frame. The authors interpret this as evidence that the CNN is learning scene texture rather than a generic rain signature, and conclude that rain detection for general-purpose surveillance cameras is not yet solved. The paper's contribution therefore is a public benchmark (VIRADA), an open-source reimplementation of the baseline, and a careful demonstration of where the data-driven method wins and where it breaks.

Load-bearing premise

The ground truth assumes that the laser disdrometer, situated roughly one kilometer from each camera, records the same rain that is falling at the camera scene, so per-minute labels are accurate for every frame of video.

Editorial extensions

If this is right

  • Deploying this detector as a preprocessing stage for rain-removal algorithms is feasible only at sites whose surfaces resemble the training data, or after fine-tuning on the target scene.
  • The VIRADA dataset gives the community a shared 215-hour benchmark with laser-disdrometer labels, so future rain-detection methods can be compared on identical surveillance footage.
  • For general-purpose camera networks, the practical implication is that a site-specific region of interest (e.g., a brick wall) is a safer choice than a textured surface like asphalt.
  • The comparison quantifies how far hand-crafted streak-orientation methods lag behind learned features on this task.

Reading between the lines

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

  • The failure on asphalt hints that the network may be short-cutting the task by recognizing static scene appearance rather than temporal rain dynamics; training on a larger variety of surfaces (or adding synthetic rain) could be a testable extension.
  • Because the camera and disdrometer are separated by about a kilometer, some per-minute label mismatches are inevitable; a denser network of local sensors could sharpen both training and evaluation.
  • The same fully-convolutional architecture could be repurposed to estimate rain rate rather than binary presence, using camera geometry approaches, if labeled rainfall volume were available.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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 presents the AAU VIRADA dataset, 215 hours of surveillance video from two traffic crossings with rainfall ground truth from a laser disdrometer and a tipping-bucket rain gauge, and compares a reimplementation of the Bossu et al. rain detection method with a 3D convolutional network (C3D) converted to a fully convolutional network. The authors report that the proposed C3D method outperforms Bossu et al. on all metrics for both crossings, and also analyze how the choice of region of interest affects generalization. The dataset and implementations are released publicly.

Significance. If the central comparative claim were supported, this would be a useful benchmark and baseline for rain detection from general-purpose surveillance cameras, and the public dataset plus open-source implementations would be a valuable community resource. The paper's strengths are the scale of the dataset, the explicit use of MCC as a primary metric for imbalanced data, and the public release of code and data. However, the headline claim is contradicted by the paper's own Table 5, and the evaluation protocol contains circularity and label-quality issues that currently undermine the reported comparisons.

major comments (4)
  1. [Abstract and Section 6] The abstract states that the proposed 3D CNN 'outperforms the previous state-of-the-art method by a large margin on all metrics, for both of the traffic crossings,' but Table 5 contradicts this. On Crossing2-asphalt, C3D-FCN achieves F1=0.0 and MCC=0.0, while Bossu-EM achieves F1=0.136 and MCC=0.008, so Bossu is better on two of the three metrics. The body text acknowledges this failure in Section 6 ('the C3D CNN evaluating the entire frame predicts no rain'), so the abstract and contribution statements must be revised to match the actual results, which support only the more nuanced conclusion in Section 7.
  2. [Section 5.1 and Table 3] The Bossu parameters are selected by a search on six video snippets from Crossing1 and then the method is evaluated on the full Crossing1-trn and Crossing1-val splits. If those snippets are part of the evaluation sets, the hyperparameters are tuned on the test data, which invalidates the Crossing1 comparison as an estimate of generalization. The authors should clarify whether the tuning snippets are included in the reported splits and should evaluate Bossu on a held-out set not used for parameter selection.
  3. [Section 5.2 and Table 5] Table 5 reports C3D-FCN and C3D-Center results on Crossing1-trn, which is the training split used to fit the network. Reporting performance on the training set is not evidence of generalization and inflates the apparent superiority of the C3D method. The comparison should focus on Crossing1-val and the Crossing2-test splits, with the training row either removed or clearly labeled as training fit rather than as a benchmark result.
  4. [Section 3.2 and Table 1] The ground-truth rain labels come from a laser disdrometer located 1230 m (Crossing1) and 970 m (Crossing2) from the cameras, and per-minute gauge readings are converted to per-frame labels. Rainfall is spatially variable at these distances, so the labels may not match the rain condition at the camera scene. This label noise affects every reported metric and should be quantified, for example by comparing the disdrometer and tipping-bucket records, by analyzing the sensitivity of results to label temporal smoothing, or by restricting evaluation to periods of spatially homogeneous rain.
minor comments (5)
  1. [Throughout] There are several typographical errors, including 'at-glace' in the Figure 1 caption, 'millenium' in Section 2, 'trough' in Section 3.1, and 'Kolomogrov' in Section 4.1; these should be corrected.
  2. [Section 6] In the text, C3D-Center on Crossing2-asphalt is reported as having an MCC of -0.08, while Table 5 reports -0.0837; the rounding should be consistent.
  3. [Section 5.2] The paper says the network is trained with a temporal stride of 8 frames and that the method 'analyse 16 frames at a time'; it would be clearer to state explicitly how the 16-frame clips are formed relative to the stride and the per-frame label conversion.
  4. [Section 6] The phrase 'outperforms the Bossu algorithm on all of the Crossing1 videos' is imprecise because the training split is included; it should be clarified that the strong result is on Crossing1-val or qualified as training-set performance.
  5. [Table 5] The table would benefit from a column showing the number of predictions for each method, since the C3D methods produce far fewer predictions than the per-frame Bossu methods; this is explained in the text but should be visible in the table.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the empirical comparison is self-contained; the abstract overclaim on Crossing2-asphalt is a factual inconsistency, not a circular derivation.

full rationale

This paper is an empirical benchmark rather than a derivation chain, so the circularity patterns are largely inapplicable. The central comparison is based on held-out validation data (Crossing1-val) and an unseen crossing (Crossing2, especially the brick crop), where the C3D model was not trained and the Bossu parameters were not tuned. The only in-sample reporting is the Crossing1-trn row, where Bossu parameters were selected using six snippets from Crossing1 (Section 5.1: "we perform a parameter search on six video snippets from the Crossing1 dataset") and the C3D network was trained on Crossing1-trn (Section 5.2). The paper explicitly acknowledges this overlap in Section 6: "As we trained the C3D CNN on a subset of the Crossing1 dataset and determined the parameters of the Bossu algorithm on the very same dataset, the difference in performance is striking." This is a methodological weakness in using training-set metrics as evidence, but it is not a circular reduction: the Crossing1-val and Crossing2-brick results are independent of the fitted parameters, and they still show C3D outperforming Bossu. The abstract's claim that C3D "outperforms the previous state-of-the-art method by a large margin on all metrics, for both of the traffic crossings" is contradicted by Table 5 on Crossing2-asphalt, where C3D-FCN scores F1=0.0 and MCC=0.0 while Bossu-EM scores F1=0.136 and MCC=0.008. That is an internal factual inconsistency, not a circular step. The paper's own limitations in Section 6 ("The C3D CNN evaluating the entire frame predicts no rain") and Section 7 ("our 3D CNN failed to function") further undercut the abstract but do not create circularity. There is no load-bearing self-citation, no imported uniqueness theorem, and no ansatz smuggled in via citation. Therefore, no circularity is present.

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

The main free parameters are the Bossu thresholds, tuned on a subset of the same dataset used for evaluation. The key domain assumption is that distant rain gauges provide accurate frame labels. No new physical entities are introduced.

free parameters (5)
  • c (photometric constraint threshold) = 3
    Selected via a grid search over the values [3, 5] on six Crossing1 snippets (Table 3).
  • Maximum BLOB size = 200
    Selected from the range [50:50:200] during the Bossu parameter search (Table 3).
  • dm (orientation uncertainty scaling) = 0.50
    Selected from [0.5:0.5:2.0] in the parameter search (Table 3).
  • Dc (KS goodness-of-fit threshold) = 0.19
    Selected from [0.01:0.01:0.20] in the parameter search (Table 3).
  • Pi_rain (Gaussian ratio threshold) = 0.40
    Selected from [0.20:0.02:0.50] in the parameter search (Table 3).
assumptions (4)
  • domain assumption Laser disdrometer measurements at 970-1230 m from the camera represent rainfall at the camera scene.
    Used to label all frames in Section 3.2; spatially variable rainfall may make these labels inaccurate.
  • domain assumption Rain streaks are visible and detectable in general-purpose surveillance video with the given camera settings.
    The detection task presupposes that rain appears as detectable streaks or patterns in the imagery (Sections 1 and 2).
  • domain assumption The photometric constraint (Eq. 1-2) and the Gaussian-uniform HOS model (Eq. 7) are valid for rain streak segmentation and detection in the Bossu method.
    Adopted from Bossu et al. [6] in Section 4.1.
  • domain assumption A 16-frame clip with stride 8 captures enough temporal rain information for classification by C3D.
    Training setup described in Section 5.2.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Is it Raining Outside? Detection of Rainfall using General-Purpose Surveillance Cameras." pith.science (2026). https://pith.science/paper/5EHMMII7

@misc{pith2026190804034,
  author       = {Pith},
  title        = {Pith review of: Is it Raining Outside? Detection of Rainfall using General-Purpose Surveillance Cameras},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5EHMMII7}},
  note         = {Machine review of arXiv:1908.04034}
}
read the original abstract

In integrated surveillance systems based on visual cameras, the mitigation of adverse weather conditions is an active research topic. Within this field, rain removal algorithms have been developed that artificially remove rain streaks from images or video. In order to deploy such rain removal algorithms in a surveillance setting, one must detect if rain is present in the scene. In this paper, we design a system for the detection of rainfall by the use of surveillance cameras. We reimplement the former state-of-the-art method for rain detection and compare it against a modern CNN-based method by utilizing 3D convolutions. The two methods are evaluated on our new AAU Visual Rain Dataset (VIRADA) that consists of 215 hours of general-purpose surveillance video from two traffic crossings. The results show that the proposed 3D CNN outperforms the previous state-of-the-art method by a large margin on all metrics, for both of the traffic crossings. Finally, it is shown that the choice of region-of-interest has a large influence on performance when trying to generalize the investigated methods. The AAU VIRADA dataset and our implementation of the two rain detection algorithms are publicly available at https://bitbucket.org/aauvap/aau-virada.

Figures

Figures reproduced from arXiv: 1908.04034 by the authors.

Figure 1
Figure 1. The proposed system at-a-glace. For rain removal algo [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Sample views of the traffic crossings from the AAU [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Activity diagram of the rain detection algorithm by [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Overview of the modified C3D CNN architecture. 3D conv and 2D conv denotes 3D and conventional 2D convolutions, respec [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Average accuracy per epoch for the trained C3D CNN. [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Average loss per epoch for the trained C3D CNN. [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

37 extracted references · 36 canonical work pages

  1. [1]

    Toward the camera rain gauge

    P Allamano, A Croci, and F Laio. Toward the camera rain gauge. Water Resources Research, 51(3):1744–1757, 2015. 2, 3

    work page 2015

  2. [2]

    Context-aware fusion of rgb and thermal imagery for traffic monitoring

    Thiemo Alldieck, Chris Bahnsen, and Thomas Moeslund. Context-aware fusion of rgb and thermal imagery for traffic monitoring. Sensors, 16(11):1947, 2016. 1

    work page 1947

  3. [3]

    Rain removal in traffic surveillance: Does it matter? IEEE Transactions on Intelligent Transportation Systems, 2018

    Chris H Bahnsen and Thomas B Moeslund. Rain removal in traffic surveillance: Does it matter? IEEE Transactions on Intelligent Transportation Systems, 2018. 1, 2

    work page 2018

  4. [4]

    Analysis of rain and snow in frequency space

    Peter C Barnum, Srinivasa Narasimhan, and Takeo Kanade. Analysis of rain and snow in frequency space. International journal of computer vision, 86(2-3):256, 2010. 2

    work page 2010

  5. [5]

    Utilisation d’un mod `ele probabiliste d’orientation de seg- ments pour d ´etecter des hydrom ´et´eores dans des s ´equences vid´eo

    J ´er´emie Bossu, Nicolas Hautiere, and Jean-Philippe TAREL. Utilisation d’un mod `ele probabiliste d’orientation de seg- ments pour d ´etecter des hydrom ´et´eores dans des s ´equences vid´eo. In XXIIe colloque GRETSI (traitement du signal et des images), Dijon (FRA), 8-11 septembre 2009 . GRETSI, Groupe dEtudes du Traitement du Signal et des Images,

    work page 2009

  6. [6]

    Rain or snow detection in image sequences through use of a histogram of orientation of streaks

    J ´er´emie Bossu, Nicolas Hauti `ere, and Jean-Philippe Tarel. Rain or snow detection in image sequences through use of a histogram of orientation of streaks. International Journal of Computer Vision, 93(3):348–367, Jul 2011. 2, 3, 4, 5

    work page 2011

  7. [7]

    A re- view of computer vision techniques for the analysis of urban traffic

    Norbert Buch, Sergio A Velastin, and James Orwell. A re- view of computer vision techniques for the analysis of urban traffic. IEEE Transactions on Intelligent Transportation Sys- tems, 12(3):920–939, 2011. 1

    work page 2011

  8. [8]

    Nvvl: Nvidia video loader

    Jared Casper, Jon Barker, and Bryan Catanzaro. Nvvl: Nvidia video loader. https://github.com/NVIDIA/ nvvl, 2018. 6

    work page 2018

Show all 37 references
  1. [9]

    Error- optimized sparse representation for single image rain re- moval

    Bo-Hao Chen, Shih-Chia Huang, and Sy-Yen Kuo. Error- optimized sparse representation for single image rain re- moval. Industrial Electronics, IEEE Transactions on , 64(8):6573–6581, 2017. 2

    2017

  2. [10]

    Ten quick tips for machine learning in com- putational biology

    Davide Chicco. Ten quick tips for machine learning in com- putational biology. BioData Mining, 10(1):35, Dec 2017. 7

    2017

  3. [11]

    Measurements of rainfall rates from videos

    Rong Dong, Juan Liao, Bo Li, Huiyu Zhou, and Danny Crookes. Measurements of rainfall rates from videos. In 2017 10th International Congress on Image and Signal Pro- cessing, BioMedical Engineering and Informatics (CISP- BMEI), pages 1–9. IEEE, 2017. 2, 3

    2017

  4. [12]

    Photometric model of a rain drop

    Kshitiz Garg and Shree K Nayar. Photometric model of a rain drop. In CMU Technical Report. 2003. 2

    2003

  5. [13]

    Kshitiz Garg and Shree K. Nayar. Vision and rain. Inter- national Journal of Computer Vision, 75(1):3–27, Oct. 2007. 2

    2007

  6. [14]

    Sam- pling errors of tipping-bucket rain gauge measurements

    Emad Habib, Witold F Krajewski, and Anton Kruger. Sam- pling errors of tipping-bucket rain gauge measurements. Journal of Hydrologic Engineering, 6(2):159–166, 2001. 3

    2001

  7. [15]

    Kensho Hara, Hirokatsu Kataoka, and Yutaka Satoh. Can spatiotemporal 3d cnns retrace the history of 2d cnns and imagenet? In Proceedings of the IEEE Conference on Com- puter Vision and Pattern Recognition (CVPR), pages 6546– 6555, 2018. 5

    2018

  8. [16]

    Meteorological conditions processing for vision-based traffic monitoring

    Nicolas Hautiere, Erwan Bigorgne, J ´er´emie Bossu, and Didier Aubert. Meteorological conditions processing for vision-based traffic monitoring. In The Eighth International Workshop on Visual Surveillance-VS2008, 2008. 2

    2008

  9. [17]

    Advancing opportunistic sensing in hydrology: a novel ap- proach to measuring rainfall with ordinary surveillance cam- eras

    Shijie Jiang, Vladan Babovic, Yi Zheng, and Jianzhi Xiong. Advancing opportunistic sensing in hydrology: a novel ap- proach to measuring rainfall with ordinary surveillance cam- eras. Water Resources Research, 2019. 3

    2019

  10. [18]

    A novel tensor-based video rain streaks removal approach via utilizing discriminatively in- trinsic priors

    Tai-Xiang Jiang, Ting-Zhu Huang, Xi-Le Zhao, Liang-Jian Deng, and Yao Wang. A novel tensor-based video rain streaks removal approach via utilizing discriminatively in- trinsic priors. In Computer Vision and Pattern Recognition, IEEE Conference on, 2017. 2

    2017

  11. [19]

    Auto- matic single-image-based rain streaks removal via image decomposition

    Li-Wei Kang, Chia-Wen Lin, and Yu-Hsiang Fu. Auto- matic single-image-based rain streaks removal via image decomposition. IEEE Transactions on Image Processing , 21(4):1742–1755, 2012. 2

    2012

  12. [20]

    Video deraining and desnowing using temporal correlation and low-rank matrix completion

    Jin-Hwan Kim, Jae-Young Sim, and Chang-Su Kim. Video deraining and desnowing using temporal correlation and low-rank matrix completion. IEEE Transactions on Image Processing, 24(9):2658–2670, 2015. 2

    2015

  13. [21]

    Sin- gle image deraining: A comprehensive benchmark analysis

    Siyuan Li, Iago Breno Araujo, Wenqi Ren, Zhangyang Wang, Eric K Tokuda, Roberto Hirata Junior, Roberto Cesar- Junior, Jiawan Zhang, Xiaojie Guo, and Xiaochun Cao. Sin- gle image deraining: A comprehensive benchmark analysis. arXiv preprint arXiv:1903.08558, 2019. 1

    1903 arXiv

  14. [22]

    Erase or fill? deep joint recurrent rain removal and recon- struction in videos

    Jiaying Liu, Wenhan Yang, Shuai Yang, and Zongming Guo. Erase or fill? deep joint recurrent rain removal and recon- struction in videos. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition , pages 3233– 3242, 2018. 2

    2018

  15. [23]

    Pixel based temporal analysis using chromatic property for remov- ing rain from videos

    Peng Liu, Jing Xu, Jiafeng Liu, and Xianglong Tang. Pixel based temporal analysis using chromatic property for remov- ing rain from videos. Computer and Information Science , 2(1):53, 2009. 2

    2009

  16. [24]

    An optical disdrometer for measuring size and velocity of hydrometeors

    Martin L ¨offler-Mang and J ¨urg Joss. An optical disdrometer for measuring size and velocity of hydrometeors. Journal of Atmospheric and Oceanic Technology, 17(2):130–139, 2000. 3

    2000

  17. [25]

    Madsen, P.S

    H. Madsen, P.S. Mikkelsen, D. Rosbjerg, and P. Harremos. Estimation of regional intensity-duration-frequency curves for extreme precipitation. Water Science and Technology , 37(11):29 – 36, 1998. Use of Historical Rainfall Series for Hydrological Modelling. 3

    1998

  18. [26]

    Mikkelsen, H

    P.S. Mikkelsen, H. Madsen, K. Arnbjerg-Nielsen, H.K. Jr- gensen, D. Rosbjerg, and P. Harremos. A rationale for using local and regional point rainfall data for design and analysis of urban storm drainage systems. Water Science and Tech- nology, 37(11):7 – 14, 1998. Use of Hist...

    1998

  19. [27]

    Rasmussen

    Jesper Ellerbæk Nielsen, Søren Liedtke Thorndahl, and Michael R. Rasmussen. Improving weather radar precipi- tation estimates by combining two types of radars. Atmo- spheric Research, 139(March):3645, 2014. 3

    2014

  20. [28]

    Rasmussen

    Jesper Ellerbæk Nielsen, Søren Liedtke Thorndahl, and Michael R. Rasmussen. A numerical method to generate high temporal resolution precipitation time series by com- bining weather radar measurements with a nowcast model. Atmospheric Research, 138(March):1–12, 2014. 3

    2014

  21. [29]

    Automatic differentiation in pytorch

    Adam Paszke, Sam Gross, Soumith Chintala, Gregory Chanan, Edward Yang, Zachary DeVito, Zeming Lin, Al- ban Desmaison, Luca Antiga, and Adam Lerer. Automatic differentiation in pytorch. In NIPS-W, 2017. 6

    2017

  22. [30]

    Video desnowing and deraining based on ma- trix decomposition

    Weihong Ren, Jiandong Tian, Zhi Han, Antoni Chan, and Yandong Tang. Video desnowing and deraining based on ma- trix decomposition. In Computer Vision and Pattern Recog- nition, IEEE Conference on, July 2017. 2

    2017

  23. [31]

    Utilizing local phase information to remove rain from video

    Varun Santhaseelan and Vijayan K Asari. Utilizing local phase information to remove rain from video. International Journal of Computer Vision, 112(1):71–89, 2015. 2

    2015

  24. [32]

    Chris Stauffer and W. Eric L. Grimson. Learning patterns of activity using real-time tracking. IEEE Transactions on pattern analysis and machine intelligence , 22(8):747–757,

  25. [33]

    Learning spatiotemporal features with 3d convolutional networks

    Du Tran, Lubomir Bourdev, Rob Fergus, Lorenzo Torresani, and Manohar Paluri. Learning spatiotemporal features with 3d convolutional networks. In Proceedings of the 2015 IEEE International Conference on Computer Vision (ICCV), ICCV ’15, pages 4489–4497, Washington, DC, USA, 201...

    2015

  26. [34]

    D. Tran, H. Wang, L. Torresani, J. Ray, Y . LeCun, and M. Paluri. A closer look at spatiotemporal convolutions for ac- tion recognition. In 2018 IEEE/CVF Conference on Com- puter Vision and Pattern Recognition , pages 6450–6459, June 2018. 5

    2018

  27. [35]

    Video post processing: low-latency spatiotemporal approach for detection and re- moval of rain

    AK Tripathi and S Mukhopadhyay. Video post processing: low-latency spatiotemporal approach for detection and re- moval of rain. IET Image Processing, 6(2):181–196, 2012. 2

    2012

  28. [36]

    Non-local neural networks

    Xiaolong Wang, Ross Girshick, Abhinav Gupta, and Kaim- ing He. Non-local neural networks. CVPR, 2018. 5

    2018

  29. [37]

    Deep joint rain detection and removal from a single image

    Wenhan Yang, Robby T Tan, Jiashi Feng, Jiaying Liu, Zong- ming Guo, and Shuicheng Yan. Deep joint rain detection and removal from a single image. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recog- nition, pages 1357–1366, 2017. 2

    2017

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.