Albumentations: Fast and Flexible Image Augmentations
Open accessThe DOI matches its Crossref record; 67 of the 67 references it cites were found in a registry or an index.
Abstract
Data augmentation is a commonly used technique for increasing both the size and the diversity of labeled training sets by leveraging input transformations that preserve corresponding output labels. In computer vision, image augmentations have become a common implicit regularization technique to combat overfitting in deep learning models and are ubiquitously used to improve performance. While most deep learning frameworks implement basic image transformations, the list is typically limited to some variations of flipping, rotating, scaling, and cropping. Moreover, image processing speed varies in existing image augmentation libraries. We present Albumentations, a fast and flexible open source library for image augmentation with many various image transform operations available that is also an easy-to-use wrapper around other augmentation libraries. We discuss the design principles that drove the implementation of Albumentations and give an overview of the key features and distinct capabilities. Finally, we provide examples of image augmentations for different computer vision tasks and demonstrate that Albumentations is faster than other commonly used image augmentation tools on most image transform operations.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“The data set used to build the diagnostic model is the ISBI 2019 data set [16–20], and 10661 cell pictures of 73 subjects were selected, including 7272 pictures of leukemic B-lymphoblast cells (cancer cells) from 47 all patients and 3389 pictures of B-lymphoid precursors (normal cells) from 26 healthy persons.”
From Method for diagnosis of acute lymphoblastic leukemia based on vit-cnn ensemble model (2021) · cites this paper as [20]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155 · cited 1×
“In computer vision, ImageNet database and the Large Scale Visual Recognition Challenge [23] have become the main platforms for benchmarking modern deep learning models for image classification after the breakthrough performance of AlexNet in 2012 that nearly halved previous state-of-the-art error rates [24].”
MethodU-Net: Convolutional Networks for Biomedical Image Segmentation
Ronneberger, O., Fischer, P., & Brox, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24574-4_28 · Cited by 99,404
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333 · cited 1×
“In computer vision, ImageNet database and the Large Scale Visual Recognition Challenge [23] have become the main platforms for benchmarking modern deep learning models for image classification after the breakthrough performance of AlexNet in 2012 that nearly halved previous state-of-the-art error rates [24].”
MethodGradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805 · cited 1×
“Basic image augmentations have been identified as an essential technique for training neural networks for visual recognition before that [25], and AlexNet extended their use to random crops, translations, horizontal flips, and altering the intensities of RGB channels to achieve that seminal result.”
BackgroundYou Only Look Once: Unified, Real-Time Object Detection
Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.91 · Cited by 53,803 · cited 2×
“Albumentations supports seamless augmentation of bounding boxes in three most common formats: COCO [59], PascalVOC [60], and YOLO [61].”
MethodGoing deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562
Microsoft COCO: Common Objects in Context
Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., & Zitnick, C. L. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10602-1_48 · Cited by 45,460
PyTorch: An Imperative Style, High-Performance Deep Learning Library
Paszke, A., Gross, S., Massa, F., Lerer, A., Bradbury, J., Chanan, G., Killeen, T., Lin, Z., Gimelshein, N., Antiga, L., et al. (2019) · arXiv · doi:10.48550/arxiv.1912.01703 · Cited by 31,473 · cited 5×
“To support users of different deep learning frameworks, Albumentations provides a convenient Python interface to enable a seamless integration with PyTorch [13], Keras [12], and Tensorflow [11].”
BackgroundMethodInfluentialA survey on deep learning in medical image analysis
Litjens, G., Kooi, T., Bejnordi, B. E., Setio, A. A. A., Ciompi, F., Ghafoorian, M., van der Laak, J. A. W. M., van Ginneken, B., & Sánchez, C. I. (2017) · Medical Image Analysis · doi:10.1016/j.media.2017.07.005 · Cited by 15,332 · cited 1×
“One typical use case for keypoint annotations is image registration, a common task in biomedical image analysis [29,63].”
Background
Research works printed but not linked: 5
- Zhang, H., Cisse, M., Dauphin, Y.N., and Lopez-Paz, D. (May, January 30). mixup: Beyond empirical risk minimization. Proceedings of the International Conference on Learning Representations (ICLR), Vancouver, BC, Canada.
- Lee, H., Hwang, S.J., and Shin, J. (2019). Rethinking Data Augmentation: Self-Supervision and Self-Distillation. arXiv.
- Lim, S., Kim, I., Kim, T., Kim, C., and Kim, S. (2019, January 8–14). Fast AutoAugment. Proceedings of the Advances in Neural Information Processing Systems (NeurIPS), Vancouver, BC, Canada.
- Brito, J.J., Li, J., Moore, J.H., Greene, C.S., Nogoy, N.A., Garmire, L.X., and Mangul, S. (2020). Enhancing rigor and reproducibility by improving software availability, usability, and archival stability. arXiv.
- Rakhlin, A., Tiulpin, A., Shvets, A.A., Kalinin, A.A., Iglovikov, V.I., and Nikolenko, S. (2019, January 16–20). Breast tumor cellularity assessment using deep neural networks. Proceedings of the IEEE International Conference on Computer Vision (ICCV) Workshops, Long Beach, CA, USA. · doi:10.1101/628693
Cited by
Items 201–300 of 1,006
A dense residual U-net for multiple sclerosis lesions segmentation from multi-sequence 3D MR images
Sarica, B., Şeker, D. Z., & Bayram, B. (2022) · International Journal of Medical Informatics · doi:10.1016/j.ijmedinf.2022.104965 · Cited by 59
GDALR: Global Dual Attention and Local Representations in transformer for surface defect detection
Zhou, X., Zhou, S., Zhang, Y., Ren, Z., Jiang, Z., & Luo, H. (2024) · Measurement · doi:10.1016/j.measurement.2024.114398 · Cited by 53
Yang, X., She, H., Lou, M., Ye, H., Guan, J., Li, J., Xiang, Z., Shen, H., & Zhang, B. (2024) · IEEE Transactions on Automation Science and Engineering · doi:10.1109/tase.2024.3375300 · Cited by 52
Medical Image Segmentation Using Deep Neural Networks with Pre-trained Encoders
Kalinin, A. A., Iglovikov, V., Rakhlin, A., & Shvets, A. A. (2020) · Advances in intelligent systems and computing · doi:10.1007/978-981-15-1816-4_3 · Cited by 50
Self-citationAVN: An Adversarial Variation Network Model for Handwritten Signature Verification
Li, H., Wei, P., & Hu, P. (2021) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2021.3056217 · Cited by 57
Tam, A. Y.-C., So, B. P.-H., Chan, T. T.-C., Cheung, A. K.-Y., Wong, D. W., & Cheung, J. C. (2021) · Sensors · doi:10.3390/s21165553 · Cited by 53
Illarionova, S., Shadrin, D., Ignatiev, V., Shayakhmetov, S., Trekin, A., & Oseledets, I. V. (2022) · IEEE Access · doi:10.1109/access.2022.3161568 · Cited by 55
Warp Consistency for Unsupervised Learning of Dense Correspondences
Truong, P., Danelljan, M., Yu, F., & Gool, L. V. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.01018 · Cited by 64
Pombo, G., Gray, R., Cardoso, M. J., Ourselin, S., Rees, G. E., Ashburner, J., & Nachev, P. (2022) · Medical Image Analysis · doi:10.1016/j.media.2022.102723 · Cited by 76
Feature pyramid network with self-guided attention refinement module for crack segmentation
Ong, J. C., Lau, S. L. H., Ismadi, M., & Wang, X. (2022) · Structural Health Monitoring · doi:10.1177/14759217221089571 · Cited by 47
“59 The network is trained on the Google Colaboratory platform.”
MethodBolon, I., Picek, L., Durso, A. M., Alcoba, G., Chappuis, F., & de Castañeda, R. R. (2022) · PLoS neglected tropical diseases · doi:10.1371/journal.pntd.0010647 · Cited by 51
Automatic monitoring of lettuce fresh weight by multi-modal fusion based deep learning
Lin, Z., Fu, R., Ren, G., Zhong, R., Ying, Y. B., & Lin, T. (2022) · Frontiers in Plant Science · doi:10.3389/fpls.2022.980581 · Cited by 51
Romero, D. J. D., den Eynde, S. V., Zaplana, I., Zhou, C., Sterkens, W., Goedemé, T., & Peeters, J. R. (2023) · Resources Conservation and Recycling · doi:10.1016/j.resconrec.2023.106865 · Cited by 54
Khan, A., Brouwer, N. P. M., Blank, A., Müller, F. C., Soldini, D., Noske, A., Gaus, E., Brandt, S., Nagtegaal, I. D., Dawson, H., et al. (2023) · Modern Pathology · doi:10.1016/j.modpat.2023.100118 · Cited by 52
Hettihewa, K. G., Kobchaisawat, T., Tanpowpong, N., & Chalidabhongse, T. H. (2023) · Scientific Reports · doi:10.1038/s41598-023-46580-4 · Cited by 53
Advanced computer vision for extracting georeferenced vehicle trajectories from drone imagery
Fónod, R., Cho, H., Yeo, H., & Geroliminis, N. (2025) · Transportation Research Part C Emerging Technologies · doi:10.1016/j.trc.2025.105205 · Cited by 64
“Furthermore, the Albumentations framework [125] was employed to introduce additional perturbations, including Gaussian and median blur, grayscale conversion, and CLAHE, each applied with a probability of 0.01, to improve robustness against variations in texture and illumination.”
InfluentialBadgujar, C., Armstrong, P. R., Gerken, A. R., Pordesimo, L. O., & Campbell, J. F. (2023) · Journal of Stored Products Research · doi:10.1016/j.jspr.2023.102196 · Cited by 50
Alrfou, K., Zhao, T., & Kordijazi, A. (2024) · Integrating materials and manufacturing innovation · doi:10.1007/s40192-024-00369-z · Cited by 62
“Implementations of data augmentation include Albumenta-tions library [114], Tensorflow-Keras, and generative models [115].”
Local climate zone classification using a multi-scale, multi-level attention network
Kim, M., Jeong, D., & Kim, Y. (2021) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2021.09.015 · Cited by 51
Cross-Forgery Analysis of Vision Transformers and CNNs for Deepfake Image Detection
Coccomini, D. A., Caldelli, R., Falchi, F., Gennaro, C., & Amato, G. (2022) · Proceedings of the 1st International Workshop on Multimedia AI against Disinformation · doi:10.1145/3512732.3533582 · Cited by 47
“We used the Albumentations library [1], and during the training, we randomly applied common transformations.”
MethodViTALnet: Anomaly on Industrial Textured Surfaces With Hybrid Transformer
Tao, X., Adak, C., Chun, P., Yan, S., & Liu, H. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3250225 · Cited by 59
“9) Noise in the Training Samples : In this experiment, the training samples were processed by the data augmentation library Albumentations [34] to simulate noise in the images, e.g., blurring, random brightness/contrast adjustments, and Gaussian noise addition.”
MethodDetection and Classification of Printed Circuit Boards Using YOLO Algorithm
Glučina, M., Anđelić, N., Lorencin, I., & Car, Z. (2023) · Electronics · doi:10.3390/electronics12030667 · Cited by 50
GenConViT: Deepfake Video Detection Using Generative Convolutional Vision Transformer
Deressa, D. W., Mareen, H., Lambert, P., Atnafu, S., Akhtar, Z., & Van Wallendael, G. (2025) · Applied Sciences · doi:10.3390/app15126622 · Cited by 46
“The Albumentation [5] library was used for data augmentation, and the following augmentation techniques were used with a strong augmentation rate of 90% : Random-Rotate, Transpose, HorizontalFlip, VerticalFlip, GaussNoise, ShiftScaleRotate, CLAHE, Sharpen, IAAEmboss, Random-BrightnessContrast, and…”
Kang, K.-S., Cho, Y.-W., Jin, K., Kim, Y.-B., & Ryu, H.-G. (2021) · Automation in Construction · doi:10.1016/j.autcon.2021.104034 · Cited by 53
Predicting biochemical recurrence of prostate cancer with artificial intelligence
Pinckaers, H., van Ipenburg, J., Melamed, J., Marzo, A. M. D., Platz, E. A., van Ginneken, B., van der Laak, J. A. W. M., & Litjens, G. J. S. (2022) · Communications Medicine · doi:10.1038/s43856-022-00126-3 · Cited by 54
Autonomous Image-Based Corrosion Detection in Steel Structures Using Deep Learning
Das, A., Dorafshan, S., & Kaabouch, N. (2024) · Sensors · doi:10.3390/s24113630 · Cited by 54
Estimating optimal crop-load for individual branches in apple tree canopies using YOLOv8
Ahmed, D., Sapkota, R., Churuvija, M., & Karkee, M. (2024) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2024.109697 · Cited by 52
Sharma, N., & Saharia, M. (2025) · Science of Remote Sensing · doi:10.1016/j.srs.2025.100203 · Cited by 61
Multi-level dilated residual network for biomedical image segmentation
Gudhe, N. R., Behravan, H., Sudah, M., Okuma, H., Vanninen, R. L., Kosma, V., & Mannermaa, A. (2021) · Scientific Reports · doi:10.1038/s41598-021-93169-w · Cited by 52
“Additionally, we applied affine, elastic, and pixel-level data augmentation techniques using Albumentations Python library 28 during the training process.”
Methodde Oliveira, Í. O., Laroca, R., Menotti, D., Fonseca, K. V. O., & Minetto, R. (2021) · IEEE Access · doi:10.1109/access.2021.3097964 · Cited by 50
“We used Albumentations [57], which is a well-known Python library for image augmentation, to apply these transformations.”
MethodLi, Y., Zeng, G., Zhang, Y., Wang, J., Jin, Q., Sun, L., Zhang, Q., Lian, Q., Qian, G., Xia, N., et al. (2021) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2021.3129245 · Cited by 54
“Our data augmentation was based on the fast and flexible image library Albumentations [56].”
MethodZech, J. R., Jaramillo, D., Altosaar, J., Popkin, C. A., & Wong, T. T. (2023) · Pediatric Radiology · doi:10.1007/s00247-023-05754-y · Cited by 44
“2 and supplementary material 4 [13, 14].”
Training CNNs in Presence of JPEG Compression: Multimedia Forensics vs Computer Vision
Mandelli, S., Bonettini, N., Bestagini, P., & Tubaro, S. (2020) · IEEE International Workshop on Information Forensics and Security · doi:10.1109/wifs49906.2020.9360903 · Cited by 54
“We resort to Albumentation [40] as our data augmentation library, while we use Pytorch [41] as Deep Learning framework.”
MethodSeason-Invariant GNSS-Denied Visual Localization for UAVs
Kinnari, J., Verdoja, F., & Kyrki, V. (2022) · IEEE Robotics and Automation Letters · doi:10.1109/lra.2022.3191038 · Cited by 56
“We perform various augmentations during training using [29].”
Meraj, T., Alosaimi, W. A., Alouffi, B., Rauf, H. T., Kumar, S. A., Damaševičius, R., & Alyami, H. (2021) · PeerJ Computer Science · doi:10.7717/peerj-cs.805 · Cited by 44
“According to Buslaev et al. (2020) , the images are transformed w.r.t. given dimensions using copping and padding, if needed.”
MethodA Survey of Computer Vision Techniques for Forest Characterization and Carbon Monitoring Tasks
Illarionova, S., Shadrin, D., Tregubova, P., Ignatiev, V., Efimov, A., Oseledets, I. V., & Burnaev, E. V. (2022) · Remote Sensing · doi:10.3390/rs14225861 · Cited by 46
Landslide Segmentation with U-Net: Evaluating Different Sampling Methods and Patch SizesPreprint
Soares, L. P., Dias, H. C., & Grohmann, C. H. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2007.06672 · Cited by 82
“The Albumentation library [Buslaev et al., 2018] was used to augment the data by using random rotations around 90◦, vertical, and horizontal flips.”
MethodCATs: Cost Aggregation Transformers for Visual CorrespondencePreprint
Cho, S., Hong, S., Jeon, S., Lee, Y., Sohn, K., & Kim, S. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.02520 · Cited by 72
“To apply data augmentation [6, 2] with predetermined probabilities to input images at random.”
MethodERNet: An Efficient and Reliable Human-Object Interaction Detection Network
Lim, J., Baskaran, V. M., Lim, J. M.-Y., Wong, K., See, J., & Tistarelli, M. (2023) · IEEE Transactions on Image Processing · doi:10.1109/tip.2022.3231528 · Cited by 56
“The pixel-level transformations [51] that we apply during training are Contrast Limited Adaptive Histogram Equalization (CLAHE), Random Brightness Contrast, and RGB Shift.”
MethodWagner, F., Eltner, A., & Maas, H. (2023) · International Journal of Applied Earth Observation and Geoinformation · doi:10.1016/j.jag.2023.103305 · Cited by 57
Yang, J. J., Matsushita, B., & Zhang, H. (2023) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2023.05.013 · Cited by 47
Few Shot Part Segmentation Reveals Compositional Logic for Industrial Anomaly Detection
Kim, S., An, S., Chikontwe, P., Kang, M., Adeli, E., Pohl, K. M., & Park, S. H. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i8.28703 · Cited by 74
Yang, Y., Yang, S., Li, C., Wang, Y., Pi, X., Lu, Y., & Wu, R. (2024) · High Voltage · doi:10.1049/hve2.12513 · Cited by 52
Ten quick tips for deep learning in biology
Lee, B. D., Gitter, A., Greene, C. S., Raschka, S., Maguire, F., Titus, A., Kessler, M. D., Lee, A. J., Chevrette, M. G., Stewart, P. A., et al. (2022) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1009803 · Cited by 50
Self-citationWagh, A., Jain, S., Mukherjee, A., Agu, E., Pedersen, P. C., Strong, D. M., Tulu, B., Lindsay, C., & Liu, Z. (2020) · IEEE Access · doi:10.1109/access.2020.3014175 · Cited by 48
“As the number of image samples in our datasets were inadequate for neural networks, a probabilistic data augmentation pipeline was implemented to generate synthetic augmentations using the albumentations library [34].”
SpheroidJ: An Open-Source Set of Tools for Spheroid Segmentation
Lacalle, D., Castro-Abril, H. A., Ranđelović, T., Domínguez, C., Heras, J., Mata, E. J., Mata, G., Méndez, Y., Pascual, V., & Ochoa, I. (2020) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2020.105837 · Cited by 44
Danish Fungi 2020 – Not Just Another Image Recognition Dataset
Picek, L., Šulc, M., Matas, J., Jeppesen, T. S., Heilmann‐Clausen, J., Lassøe, T., & Frøslev, T. G. (2022) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv51458.2022.00334 · Cited by 69
“For training, we utilized several augmentation techniques from the Albumentations library [5].”
MethodImproved U-Net based on contour prediction for efficient segmentation of rectal cancer
Li, D., Chu, X., Cui, Y., Zhao, J., Zhang, K., & Yang, X. (2021) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2021.106493 · Cited by 47
Elsherbiny, O., Zhou, L., He, Y., & Qiu, Z. (2022) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2022.107453 · Cited by 43
Bai, Y., Zha, B., Sezen, H., & Yilmaz, A. (2023) · Structural Health Monitoring · doi:10.1177/14759217221083649 · Cited by 42
FV-ViT: Vision Transformer for Finger Vein Recognition
Li, X., & Zhang, B.-B. (2023) · IEEE Access · doi:10.1109/access.2023.3297212 · Cited by 43
Abbasi, R., Martínez, P., & Ahmad, R. (2022) · Artificial Intelligence in Agriculture · doi:10.1016/j.aiia.2023.09.001 · Cited by 47
Scala, P., Manno, G., & Ciraolo, G. (2024) · Computers & Geosciences · doi:10.1016/j.cageo.2024.105704 · Cited by 42
Arrow R-CNN for handwritten diagram recognition
Schäfer, B., Keuper, M., & Stuckenschmidt, H. (2021) · International Journal on Document Analysis and Recognition (IJDAR) · doi:10.1007/s10032-020-00361-1 · Cited by 45
“We use the Albumentations library [8] for all augmentations except IAMWordAugmentation .”
MethodRice yield estimation using a CNN-based image-driven data assimilation framework
Han, J., Shi, L., Yang, Q. L., Chen, Z., Yu, J., & Zha, Y. (2022) · Field Crops Research · doi:10.1016/j.fcr.2022.108693 · Cited by 44
Zheng, T.-L., Qin, H.-Y., Cui, Y., Wang, R., Zhao, W., Zhang, S.-J., Geng, S., & Zhao, L. (2023) · BMC Medical Imaging · doi:10.1186/s12880-023-01011-8 · Cited by 38
“Data augmentation strategies include increasing and decreasing the brightness, increasing and decreasing the contract, horizontal mirroring, random angular rotation, random cropping, random stretching, etc. [42].”
BackgroundMetal Surface Defect Detection Method Based on Improved Cascade R-CNN
Wang, Y., Wang, X., Hao, R., Lu, B., & Huang, B. (2023) · Journal of Computing and Information Science in Engineering · doi:10.1115/1.4063860 · Cited by 47
Novel Category Discovery Without Forgetting for Automatic Target Recognition
Huang, H., Gao, F., Sun, J., Wang, J., Hussain, A., & Zhou, H. (2024) · IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing · doi:10.1109/jstars.2024.3358449 · Cited by 42
“This approach utilized hierarchical embedding networks and pseudoincremental strategies for effective model training.”
MethodUrban Sound Tagging using Convolutional Neural Networks
Adapa, S. (2019) · Detection and Classification of Acoustic Scenes and Events Workshop (DCASE) · doi:10.33682/8axe-9243 · Cited by 52
Jiang, Y., Ku, B., Kim, W., & Ko, H. (2020) · IEEE Geoscience and Remote Sensing Letters · doi:10.1109/lgrs.2020.3005679 · Cited by 44
Multi-Label Contrastive Learning for Abstract Visual Reasoning
Małkiński, M., & Mańdziuk, J. (2022) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3185949 · Cited by 57
“…f θ , g φ and p ω to minimize L 18: end for 4) Linear Evaluation: After the contrastive pretraining step, we discard the projection head g φ and the rule discovery network p ω , freeze the parameters of the encoder f θ and attach a simple linear scoring head s ψ with a single-output neuron.”
BackgroundChen, H.-C., Tzeng, S.-S., Hsiao, Y., Chen, R., Hung, E.-C., & Lee, O. K. (2021) · JMIR mhealth and uhealth · doi:10.2196/32444 · Cited by 45
“The input data were processed with the following steps: (1) images were resized into 256×256 using a bilinear interpolation method, (2) images were augmented using horizontal flip and randomly rotated using the albumentations method [12], and (3) five-fold cross-validation was used to estimate the performance of the m…”
MethodDilUnet: A U-net based architecture for blood vessels segmentation
Hussain, S., Guo, F., Li, W., & Shen, Z. (2022) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2022.106732 · Cited by 41
Matsui, T., Sugimori, H., Koseki, S., & Koyama, K. (2023) · Postharvest Biology and Technology · doi:10.1016/j.postharvbio.2023.112390 · Cited by 45
Satellite Video Remote Sensing for Flood Model Validation
Masafu, C., & Williams, R. D. (2024) · Water Resources Research · doi:10.1029/2023wr034545 · Cited by 42
“To implement data augmentation, we used the Albumentations library (Buslaev et al., 2020) in Python, where we applied vertical and horizontal flips, transposition, grid distortions, elastic transforms and random gamma to both the images and corresponding binary masks (Figure 4).”
MethodButtar, P. K., & Sachan, M. K. (2022) · Expert Systems with Applications · doi:10.1016/j.eswa.2022.118380 · Cited by 48
Ho, D. J., Chui, M. H., Vanderbilt, C., Jung, J., Robson, M. E., Park, C.-S., Roh, J., & Fuchs, T. J. (2022) · Journal of Pathology Informatics · doi:10.1016/j.jpi.2022.100160 · Cited by 50
“3 flips, and color jittering were used as data augmentation transformations.(33) During the first training, we trained the model from randomly initialized parameters with a learning rate of 5 × 10, a momentum of 0.”
MethodWang, Z., Yao, J., Zeng, C., Li, L., & Tan, C. (2023) · Systems · doi:10.3390/systems11070372 · Cited by 49
Zhao, X., Zhou, B., Luo, Y., Chen, L., Zhu, L., Chang, S., Fang, X., & Yao, Z. (2023) · European Radiology · doi:10.1007/s00330-023-10505-6 · Cited by 39
“…test time augmentation (TTA) technique to create multiple versions of the original image by rotating (90, 180, or 270°), scaling (0.75 to 1.25 times), and cropping (to half-side length of the original image) through a computer vision tool of Albumentations [32], which were then put into the model.”
InfluentialAutomatic Fault Mapping in Remote Optical Images and Topographic Data With Deep Learning
Mattéo, L., Manighetti, I., Tarabalka, Y., Gaucel, J., van den Ende, M., Mercier, A., Taşar, O., Girard, N., Leclerc, F., Giampetro, T., et al. (2021) · Journal of Geophysical Research Solid Earth · doi:10.1029/2020jb021269 · Cited by 48
“…data set by applying geometrical and numerical transformations, including random horizontal rotations (here we used rotations by 90°, 180°, and 270°), flip transformations, color adjustments (with changes in contrast and gamma correction), and the addition of Gaussian noise (Buslaev et al., 2020).”
MethodCai, Y., Yang, Y., Zheng, Q., Shen, Z., Shang, Y., Yin, J., & Shi, Z. (2022) · Remote Sensing · doi:10.3390/rs14010190 · Cited by 40
Kittipongdaja, P., & Siriborvornratanakul, T. (2022) · EURASIP Journal on Image and Video Processing · doi:10.1186/s13640-022-00581-x · Cited by 39
“Our data augmentation is done by a library named albu-mentations 0.2.3 [19] (https:// album entat ions. ai/), including horizontal flip ( p = 0.5), random brightness contrast ( p = 0.5), random gamma ( p = 0.5), and grid distortion ( p = 0.5, border_mode = cv2.”
MethodInfluentialDai, G., Hu, L., & Fan, J. (2022) · Computational Intelligence and Neuroscience · doi:10.1155/2022/6114061 · Cited by 39
“For this purpose, we applied different weather data enhancement techniques to the training set using the image data enhancement method of the Albumentations [29] library in Python for simulating the changes in the environment induced by different weather conditions.”
MethodA CNN-Based Self-Supervised Synthetic Aperture Radar Image Denoising Approach
Tan, S., Zhang, X., Wang, H., Yu, L., Du, Y., Yin, J., & Wu, B. (2021) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2021.3104807 · Cited by 40
“We use random brightness and contrast, vertical and horizontal flip, and random rotation operations for data augmentation [56].”
MethodFarady, I., Lin, C.-Y., & Chang, M.-C. (2023) · Journal of Intelligent Manufacturing · doi:10.1007/s10845-023-02109-0 · Cited by 41
Marrón-Esquivel, J. M., Durán-López, L., Linares-Barranco, A., & Dominguez-Morales, J. P. (2023) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2023.106856 · Cited by 40
Cross-View Object Geo-Localization in a Local Region With Satellite Imagery
Sun, Y., Ye, Y., Kang, J. S., Fernández-Beltrán, R., Feng, S., Li, X., Luo, C., Zhang, P., & Plaza, A. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3307508 · Cited by 41
The OPS-SAT case: A data-centric competition for onboard satellite image classification
Meoni, G., Märtens, M., Derksen, D., See, K., Lightheart, T., Sécher, A., Arnaud Martin, B., Rijlaarsdam, D., Fanizza, V., & Izzo, D. (2024) · Astrodynamics · doi:10.1007/s42064-023-0196-y · Cited by 40
“These augmentations were provided using the AutoAlbument function from the Albumentations library [45].”
Tatar, A., Haghighi, M., & Zeinijahromi, A. (2024) · Journal of Rock Mechanics and Geotechnical Engineering · doi:10.1016/j.jrmge.2024.02.015 · Cited by 45
MSCPNet: A Multi-Scale Convolutional Pooling Network for Maize Disease Classification
Al-Gaashani, M. S. A. M., Alkanhel, R. I., Ali, M. A. S., Muthanna, M. S. A., Aziz, A., & Muthanna, A. (2025) · IEEE Access · doi:10.1109/access.2024.3524729 · Cited by 43
Deep Learning-Based Smart IoT Health System for Blindness Detection Using Retina Images
Jaiswal, A. K., Tiwari, P., Kumar, S., Al-Rakhami, M. S., Alrashoud, M. R., & Ghoneim, A. (2021) · IEEE Access · doi:10.1109/access.2021.3078241 · Cited by 43
“We employ the Albumentations library [37] to perform augmentation, where we flipped images horizontally and vertically, image rotation to 360 ◦ , zoom images to 1.3 x and contrasting for lightning.”
MethodInfluentialBöhland, M., Tharun, L., Scherr, T., Mikut, R., Hagenmeyer, V., Thompson, L. D. �., Perner, S., & Reischl, M. (2021) · PLoS ONE · doi:10.1371/journal.pone.0257635 · Cited by 39
Forecasting with gradient boosted trees: augmentation, tuning, and cross-validation strategies
Lainder, A. D., & Wolfinger, R. D. (2022) · International Journal of Forecasting · doi:10.1016/j.ijforecast.2021.12.003 · Cited by 55
AugLy: Data Augmentations for Adversarial Robustness
Papakipos, Z., & Bitton, J. (2022) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw56347.2022.00027 · Cited by 43
“A majority of libraries focus on images [1, 14, 19] and text [8, 18, 21], however some audio [15, 20, 29] and video [6,16,32] augmentation libraries do exist as well (see Section 4 for in-depth comparisons between AugLy and existing libraries for each modality).”
BackgroundMethodXIRL: Cross-embodiment inverse reinforcement learning
Zakka, K., Zeng, A., Florence, P., Tompson, J., Bohg, J., & Dwibedi, D. (2021) · arXiv · doi:10.48550/arxiv.2106.03911 · Cited by 38
“We apply data augmentation during training using the Albumentations library [48].”
MethodLungHist700: A dataset of histological images for deep learning in pulmonary pathology
Diosdado, J., Gilabert, P., Seguí, S., & Borrego, H. (2024) · Scientific Data · doi:10.1038/s41597-024-03944-3 · Cited by 44
Deep learning detection of melanoma metastases in lymph nodes
Jansen, P., Baguer, D. O., Duschner, N., Arrastia, J. L., Schmidt, M., Landsberg, J., Wenzel, J., Schadendorf, D., Hadaschik, E. N., Maaß, P., et al. (2023) · European Journal of Cancer · doi:10.1016/j.ejca.2023.04.023 · Cited by 41
Ullah, F., Ullah, I., Khan, K., Khan, S., & Amin, F. (2025) · Applied Intelligence · doi:10.1007/s10489-024-06139-w · Cited by 46
“Highlighting the advancement in this field, the albu-mentations library, as established by [64], has emerged as a valuable resource.”
Sharan, L., Romano, G., Koehler, S., Kelm, H., Karck, M., Simone, R. D., & Engelhardt, S. (2021) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2021.3099858 · Cited by 48
Fashi, P. A., Hemati, S., Babaie, M., González, R., & Tizhoosh, H. R. (2022) · Journal of Pathology Informatics · doi:10.1016/j.jpi.2022.100133 · Cited by 45
“For data augmentation, we applied horizontal and vertical flip, 90 degree rotation, shifting, and scaling to the data from the Albumentations library.(25) We used an exponential decay learning rate scheduler with a base of 0.”
MethodGenze, N., Wirth, M., Schreiner, C., Ajekwe, R., Grieb, M., & Grimm, D. G. (2023) · Plant Methods · doi:10.1186/s13007-023-01060-8 · Cited by 44
“All models are implemented in Python 3.8.10 [53] using the packages numpy [54], pandas [55], pytorch [56], scikit-image [57], scikit-learn [58], albumentations [59] and kornia [60].”
MethodInfluentialQuality-Agnostic Deepfake Detection with Intra-model Collaborative Learning
Le, B. M., & Woo, S. S. (2023) · 2023 IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.02045 · Cited by 36
“In the test set, we include both video compression and random JPEG image compression [3].”
MethodMaxwell, A. E., Odom, W. E., Shobe, C. M., Doctor, D. H., Bester, M. S., & Ore, T. M. (2023) · Earth and Space Science · doi:10.1029/2023ea002845 · Cited by 41
“In order to further increase the variability in the training data and to potentially reduce overfitting, random changes to brightness and/or contrast and blurring were applied using the Albumentations Python library (Buslaev et al., 2020).”
MethodRapid Grapevine Health Diagnosis Based on Digital Imaging and Deep Learning
Elsherbiny, O., Elaraby, A. E., Alahmadi, M. D., Hamdan, M., & Gao, J. (2024) · Plants · doi:10.3390/plants13010135 · Cited by 43
“Augmenting data by applying a sequence of transformations—such as mirroring or rotating an image—increases the dataset’s usefulness and depth [15].”
MethodWieslander, H., Harrison, P. J., Skogberg, G., Jackson, S., Friden, M., Karlsson, J., Spjuth, O., & Wahlby, C. (2021) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2020.2996300 · Cited by 35
“The scaling and shifting factors was limited to 10% of the image size [45].”
BackgroundMaskHunter: real‐time object detection of face masks during the COVID‐19 pandemic
Cao, Z., Shao, M., Xu, L., Mu, S., & Qu, H. (2020) · IET Image Processing · doi:10.1049/iet-ipr.2020.1119 · Cited by 39
Wheat Head Detection using Deep, Semi-Supervised and Ensemble Learning
Fourati, F., Mseddi, W. S., & Attia, R. (2021) · Canadian Journal of Remote Sensing · doi:10.1080/07038992.2021.1906213 · Cited by 44
“…et al. 2019), gaussian noise injection (Moreno-Barea, Strazzera et al. 2018), and natural noise injection such as simulating the wind effect with a motion blur augmentation and simulate shadow due to the acquisition process and the position of sun using shadow augmentation (Buslaev et al. 2020).”
BackgroundMethodExploiting sample correlation for crowd counting with multi-expert network
Liu, X., Li, G., Han, Z., Zhang, W., Yang, Y., Huang, Q., & Sebe, N. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00320 · Cited by 43
“We used official division of the data set To expand the data set and applied augmentation technologies provided in [3] ( including horizontal flip, blur, random cropping and re-size, optical distortion and random contrast and brightness) and generated 6000 crops using the training data sets.”
MethodZhang, X., Han, L., Sobeih, T., Lappin, L., Lee, M. A., Howard, A., & Kisdi, Á. (2022) · Remote Sensing · doi:10.3390/rs14061400 · Cited by 39
“All models used in this paper are developed using Pytorch 1.6 and the image augmentation are based on the open-sourced image augmentation library Albumentations [67].”
BackgroundInfluentialRippner, D. A., Raja, P. V., Earles, J. M., Momayyezi, M., Buchko, A. C., Duong, F. V., Forrestel, E. J., Parkinson, D. Y., Shackel, K. A., Neyhart, J., et al. (2022) · Frontiers in Plant Science · doi:10.3389/fpls.2022.893140 · Cited by 40
“To help avoid overfitting the training data, the data was augmented using Albumentations package in Python to flip and rotate a subset of the images during model training (Buslaev et al., 2020).”
Method
Metadata
- Authors
- Buslaev, A., Iglovikov, V. I., Khvedchenya, E., Parinov, A., Druzhinin, M., & Kalinin, A. A.
- Affiliations
- Mapbox, Minsk 220030, Belarus
- Lyft (United States) (US)
- ODS.ai, Odessa 65000, Ukraine
- X5 Retail Group, Moscow 119049, Russia
- Simicon, Saint Petersburg 195009, Russia
- University of Michigan (US)
- Corresponding author
- Kalinin, Alexandr A.
- Venue
- Information
- Publisher
- MDPI AG
- Published
- 2020-02-24
- Volume
- 11
- Issue
- 2
- Pages
- 125
- ISSN
- 2078-2489 (ISSN-L)
- OpenAlex
- W2889985731
- Cited by
- 2,714
- References
- 56
- Citations by year
- 2026: 418 · 2025: 514 · 2024: 497 · 2023: 474 · 2022: 395 · 2021: 273 · 2020: 113
- FWCI
- 122.06
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Information (publishedVersion, cc-by)
- arXiv (Cornell University) (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- arXiv (Cornell University)
- Funders
- Nvidia
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and Data Classification
- Physical Sciences › Computer Science › Artificial Intelligence › Computer Science and Engineering
- Keywords
- image augmentation, Albumentations, open-source library, image transforms, data augmentation, overfitting, implicit regularization, computer vision, performance comparison
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.