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 301–400 of 1,006
Alzubaidi, M., Agus, M., Shah, U., Makhlouf, M., Alyafei, K., & Househ, M. (2022) · Diagnostics · doi:10.3390/diagnostics12092229 · Cited by 35
Promoting Single-Modal Optical Flow Network for Diverse Cross-Modal Flow Estimation
Zhou, S., Tan, W., & Yan, B. (2022) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v36i3.20268 · Cited by 59
“The data augmentations is implemented with Albu-mentations (Buslaev et al. 2020).”
MethodStandardizing and Centralizing Datasets for Efficient Training of Agricultural Deep Learning Models
Joshi, A., Guevara, D., & Earles, J. M. (2023) · Plant Phenomics · doi:10.34133/plantphenomics.0084 · Cited by 46
“We use the albumentations (Buslaev et al., 2020) library for our augmentations, noting its seamless integration into our existing PyTorch training pipeline.”
MethodGoss, D. M., Vasilescu, S. A., Vasilescu, P. A., Cooke, S., Kim, S. H., Sacks, G. P., Gardner, D. K., & Warkiani, M. E. (2024) · Reproductive BioMedicine Online · doi:10.1016/j.rbmo.2024.103910 · Cited by 36
Using improved YOLO V5s to recognize tomatoes in a continuous working environment
Gao, G., Shuai, C., Wang, S., & Ding, T. (2024) · Signal Image and Video Processing · doi:10.1007/s11760-024-03010-w · Cited by 43
“In the process of deep learning network model training, the more sufficient the sample data of the model, the stronger the generalization ability and the better the robustness of the trained network model [30–32].”
Autonomous driving in traffic with end-to-end vision-based deep learning
Paniego, S., Shinohara, E., & Cañas, J. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.127874 · Cited by 40
Universal bovine identification via depth data and deep metric learning
Sharma, A., Randewich, L., Andrew, W., Hannuna, S., Campbell, N., Mullan, S. M., Dowsey, A. W., Smith, M. L., Hansen, M., & Burghardt, T. (2024) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2024.109657 · Cited by 45
Al-Gaashani, M. S. A. M., Xu, W., & Obsie, E. Y. (2024) · Applied Soft Computing · doi:10.1016/j.asoc.2024.112553 · Cited by 35
FruitDet: Attentive Feature Aggregation for Real-Time Fruit Detection in Orchards
Kateb, F., Monowar, M. M., Hamid, M. A., Ohi, A. Q., & Mridha, M. F. (2021) · Agronomy · doi:10.3390/agronomy11122440 · Cited by 44
Feng, S., Liu, Q., Patel, A., Bazai, S. U., Jin, C., Kim, J. S., Sarrafzadeh, M., Azzollini, D., Yeoh, J., Kim, E., et al. (2022) · Journal of Medical Imaging and Radiation Oncology · doi:10.1111/1754-9485.13393 · Cited by 40
“Augmentation (resizing, horizontal and vertical flip and up to 360-degree rotations) of the training dataset was performed.(20) All candidate segmentation models were constructed to have a primary segmentation output (predicting location, size and area of pneumothorax), as measured by mean Dice and Dice coefficient fo…”
MethodWilson, R. J., de Siqueira, A. F., Brooks, S. J., Price, B. W., Simon, L., van der Walt, S., & Fenberg, P. B. (2022) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.13844 · Cited by 39
From pixels to insights: Machine learning and deep learning for bioimage analysis
Jan, M., Spangaro, A., Lenartowicz, M., & Ušaj, M. M. (2023) · BioEssays · doi:10.1002/bies.202300114 · Cited by 42
“[145,146] Sparse annotation-based training of DL models has been used for object segmentation and tracking in lieu of extensive manual image labeling with DL tools, such as ELEPHANT, StarDist, and Cellpose.”
MethodBD-SKUNet: Selective-Kernel UNets for Building Damage Assessment in High-Resolution Satellite Images
Ahmadi, S. A., Mohammadzadeh, A., Yokoya, N., & Ghorbanian, A. (2023) · Remote Sensing · doi:10.3390/rs16010182 · Cited by 29
“They can also act like regularization techniques that feed models with more challenging scenarios [58,59].”
Lee, H., Lee, J., Jang, J. H., Hwang, I., Choi, K. S., Park, J. H., Chung, J. W., & Choi, S. H. (2024) · Neuroradiology · doi:10.1007/s00234-024-03298-y · Cited by 37
“Data augmentation was conducted using the “ShiftScaleRotate” and “RandomScale” functionalities from the Python “albumentation” library [14].”
MethodAn efficient detector for detecting surface defects on cold-rolled steel strips
Chen, S., Jiang, S., Wang, X., Sun, P., Hua, C., & Sun, J. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.109325 · Cited by 43
Duan, C., & Zhang, T. (2020) · IEEE Access · doi:10.1109/access.2020.3025165 · Cited by 37
“tion, because the previous rotation makes output image larger enough to contain the entire rotated image to ensure that the pixel value beyond the image size range is not lost after the rotation of the source image [19], [20].”
BackgroundLocalization of cementoenamel junction in intraoral ultrasonographs with machine learning
Nguyen, K.-C. T., Le, B. M., Li, M., Almeida, F. T., Major, P. W., Kaipatur, N. R., Lou, E., Punithakumar, K., & Le, L. H. (2021) · Journal of Dentistry · doi:10.1016/j.jdent.2021.103752 · Cited by 34
Alirezazadeh, P., Rahimi-Ajdadi, F., Abbaspour‐Gilandeh, Y., Landwehr, N., & Tavakoli, H. (2021) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2021.106499 · Cited by 37
AutoRet: A Self-Supervised Spatial Recurrent Network for Content-Based Image Retrieval
Monowar, M. M., Hamid, M. A., Ohi, A. Q., Alassafi, M. O., & Mridha, M. F. (2022) · Sensors · doi:10.3390/s22062188 · Cited by 40
Image-based Automatic Dial Meter Reading in Unconstrained Scenarios
Salomon, G., Laroca, R., & Menotti, D. (2022) · Measurement · doi:10.1016/j.measurement.2022.112025 · Cited by 41
“While there are several data augmentation libraries available, we created the modified images using Albumentations [55], which is a wellknown Python library for image augmentation that has been successfully used in several applications, including AMR [7].”
MethodCATs++: Boosting Cost Aggregation With Convolutions and Transformers
Cho, S., Hong, S., & Kim, S. (2022) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2022.3218727 · Cited by 70
“To apply data augmentation [89], [90] with predetermined probabilities to input images at random.”
MethodMirpulatov, I., Illarionova, S., Shadrin, D., & Burnaev, E. V. (2023) · IEEE Access · doi:10.1109/access.2023.3300967 · Cited by 38
“In addition, we employ two types of augmentations, HorizontalFlip and VerticalFlip with a probability of 0.5, utilizing the Albumentations library [54].”
MethodInfluentialReview of Image Augmentation Used in Deep Learning-Based Material Microscopic Image Segmentation
Ma, J., Hu, C., Zhou, P., Jin, F., Wang, X., & Huang, H. (2023) · Applied Sciences · doi:10.3390/app13116478 · Cited by 46
Weed image augmentation by ControlNet-added stable diffusion for multi-class weed detection
Deng, B., & Lu, Y. (2025) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2025.110123 · Cited by 44
Understanding the decisions of CNNs: An in-model approach
Rio-Torto, I., Fernandes, K., & Teixeira, L. F. (2020) · Pattern Recognition Letters · doi:10.1016/j.patrec.2020.04.004 · Cited by 40
Domain-guided data augmentation for deep learning on medical imaging
Athalye, C., & Arnaout, R. (2023) · PLoS ONE · doi:10.1371/journal.pone.0282532 · Cited by 38
End-to-end Deep Learning Methods for Automated Damage Detection in Extreme Events at Various Scales
Bai, Y., Sezen, H., & Yilmaz, A. (2021) · 2020 25th International Conference on Pattern Recognition (ICPR) · doi:10.1109/icpr48806.2021.9413041 · Cited by 32
“[26] develop this method with pixel-level transformation and spatial-level transformation, including flipping, rotating, cropping, etc.”
MethodYing, W., Zheng, Q., Wu, Y., Yang, K., Zhou, Z., Chen, J., & Ye, Z. (2023) · Applied Soft Computing · doi:10.1016/j.asoc.2023.110869 · Cited by 34
Explainable convolutional neural networks for assessing head and neck cancer histopathology
Dörrich, M., Hecht, M., Fietkau, R., Hartmann, A., Iro, H., Gostian, A., Eckstein, M., & Kist, A. M. (2023) · Diagnostic Pathology · doi:10.1186/s13000-023-01407-8 · Cited by 33
“The Python library Albumen-tations [19] was used for transforming tiles and masks jointly.”
MethodProbabilistic Modeling of Inter- and Intra-observer Variability in Medical Image Segmentation
Schmidt, A., Morales-Álvarez, P., & Molina, R. (2023) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51070.2023.01929 · Cited by 39
“We use the predefined train/validation/test splits [19].”
Jooshin, H. K., Nangir, M., & Seyedarabi, H. (2024) · IET Image Processing · doi:10.1049/ipr2.13077 · Cited by 43
Improving spatial transferability of deep learning models for small-field crop yield prediction
Stiller, S., Grahmann, K., Ghazaryan, G., & Ryo, M. (2024) · ISPRS Open Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.ophoto.2024.100064 · Cited by 43
Robust object detection in extreme construction conditions
Ding, Y., Zhang, M., Pan, J., Hu, J., & Luo, X. (2024) · Automation in Construction · doi:10.1016/j.autcon.2024.105487 · Cited by 34
Distilled transformers with locally enhanced global representations for face forgery detection
Zhang, Y., Li, Q., Yu, Z., & Shen, L. (2024) · Pattern Recognition · doi:10.1016/j.patcog.2024.111253 · Cited by 40
“The Albumentations [33] library is exploited for data augmentation.”
Feature Fusion with Albumentation for Enhancing Monkeypox Detection Using Deep Learning Models
Pratama, N. R., Setiadi, D. R. I. M., Harkespan, I., & Ojugo, A. A. (2025) · Journal of Computing Theories and Applications · doi:10.62411/jcta.12255 · Cited by 35
“By providing a variety of adaptive transformations, this library allows for richer data manipulation than conventional methods while still preserving the essential characteristics of the original image[26].”
Application of neural networks to image recognition of wheat rust diseases
Александрович, Г. М., Ekaterina, S., & Afonnikov, D. A. (2020) · IEEE Ural-Siberian Conference on Computational Technologies in Cognitive Science, Genomics and Biomedicine (CSGB) · doi:10.1109/csgb51356.2020.9214703 · Cited by 33
“In the process of training, the original image was subjected to a set of the following transformations implemented in the Albumentations library[16]: HorizontalFlip(); RandomGamma(); RandomBrightnessContrast(); HueSaturationValue(); JpegCompression (); Blur ().”
MethodAbayomi‐Alli, O., Damaševičius, R., Maskeliūnas, R., & Misra, S. (2021) · Electronics · doi:10.3390/electronics10080978 · Cited by 33
“Examples of the transformation method include affine transformations, blurring, brightness shift, channel shuffle, contrast shift, elastic transformations, image blending, reflection, rotation, and scaling [28].”
MethodMixChannel: Advanced Augmentation for Multispectral Satellite Images
Illarionova, S., Nesteruk, S., Shadrin, D., Ignatiev, V., Pukalchik, M., & Oseledets, I. V. (2021) · Remote Sensing · doi:10.3390/rs13112181 · Cited by 39
“It adds variability to the data and therefore makes a model more robust [12].”
BackgroundDETECTING CRACKS AND SPALLING AUTOMATICALLY IN EXTREME EVENTS BY END-TO-END DEEP LEARNING FRAMEWORKS
Bai, Y., Sezen, H., & Yılmaz, A. (2021) · ISPRS annals of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprs-annals-v-2-2021-161-2021 · Cited by 35
“Buslaev et al. (2020) develop this method with pixel-level transformation and spatial transformation, including flipping, rotating, cropping, etc.”
InfluentialSa, I., Lim, J., Ahn, H. S., & MacDonald, B. A. (2022) · Sensors · doi:10.3390/s22134721 · Cited by 40
Kyung, S., Shin, K., Jeong, H., Kim, K. D., Park, J., Cho, K., Lee, J. H., Hong, G., & Kim, N. (2022) · Medical Image Analysis · doi:10.1016/j.media.2022.102489 · Cited by 32
Ghosh, R., & Bovolo, F. (2022) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2022.3180761 · Cited by 33
“During training, data augmentation techniques were used: i) grid distortion [38] and ii) elastic deformation [28] since they preserve the sequential structures of the signal along range.”
MethodLBPNet: Exploiting texture descriptor for deepfake detection
Kingra, S., Aggarwal, N., & Kaur, N. (2022) · Forensic Science International Digital Investigation · doi:10.1016/j.fsidi.2022.301452 · Cited by 43
Sawyer, E., Laroche-Pinel, E., Flasco, M. T., Cooper, M. L., Corrales, B., Fuchs, M., & Brillante, L. (2023) · Frontiers in Plant Science · doi:10.3389/fpls.2023.1117869 · Cited by 37
“Augmentation was performed using the Albumentations package in Python (Buslaev et al., 2020).”
MethodInfluentialKurita, Y., Meguro, S., Tsuyama, N., Kosugi, I., Enomoto, Y., Kawasaki, H., Uemura, T., Kimura, M., & Iwashita, T. (2023) · PLoS ONE · doi:10.1371/journal.pone.0285996 · Cited by 33
“During training, basic augmentations were performed using the Albumentations library [29] with the following augmentations: VerticalFlip (50%), Rotate (50%), RandomGridShuffle (50%), RandomBrightnessContrast (30%), and RandomGamma (30%).”
MethodWei, Y., Huang, W., Jian, C.-Y., Hsu, C. H., Hsu, C., Lin, C., Cheng, C., Chen, Y.-L., Wei, H., & Chen, K. (2022) · NeuroImage Clinical · doi:10.1016/j.nicl.2022.103044 · Cited by 32
“Integrating AI and clinical knowledge is a challenge for medical education (Chan and Zary, 2019), and a framework for stepwise education in professional development may shape the new generation to integrate and utilize AI-assisted information effectively (Paranjape et al., 2019).”
BackgroundRandom Interpolation Resize: A free image data augmentation method for object detection in industry
Wan, D., Lu, R., Xu, T., Shen, S., Lang, X., & Ren, Z. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.120355 · Cited by 36
Visual and tactile perception techniques for braille recognition
Park, B., Im, S., Lee, H., Lee, Y. T., Nam, C., Hong, S., & Kim, M. (2023) · Micro and Nano Systems Letters · doi:10.1186/s40486-023-00191-w · Cited by 35
“Using data augmentation techniques, such as those provided by the Albumentation library [24], these images were expanded to 520 images to construct a custom-made braille data-set.”
MethodA weak supervision machine vision detection method based on artificial defect simulation
Li, C., Huang, Y., Li, H., & Zhang, X. (2020) · Knowledge-Based Systems · doi:10.1016/j.knosys.2020.106466 · Cited by 36
Visual Identification of Articulated Object Parts
Zeng, V., Lee, T. E., Liang, J., & Kroemer, O. (2021) · IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) · doi:10.1109/iros51168.2021.9636054 · Cited by 54
“In addition to domain randomization, we apply standard image augmentation techniques during training, including geometric transformations such as random rotations, flips, and crops [41].”
MethodTian, X., Chen, C., Chen, C., Yan, Z., Wu, W., Chen, F., Chen, J., & Lv, X. (2022) · Journal of Raman Spectroscopy · doi:10.1002/jrs.6302 · Cited by 32
Deep Learning-Based Four-Region Lung Segmentation in Chest Radiography for COVID-19 Diagnosis
Kim, Y.-G., Kim, K., Wu, D., Ren, H., Tak, W. Y., Park, S. Y., Lee, Y. R., Kang, M. K., Park, J. G., Kim, B. S., et al. (2022) · Diagnostics · doi:10.3390/diagnostics12010101 · Cited by 33
“egions. However, if the model confidence of the left hilum detection is lower or equal to 0.9, the prediction box for the carina would be used. To train a robust detection model, augmentation methods [21] such as rotation, translation, shearing, scaling, pixel noise, different range of contrast, brightness, hue, and s…”
MethodInfluentialWeighted average ensemble-based semantic segmentation in biological electron microscopy images
Devan, K. S., Kestler, H. A., Read, C., & Walther, P. (2022) · Histochemistry and Cell Biology · doi:10.1007/s00418-022-02148-3 · Cited by 32
“In order to optimize computational resources while increasing dataset size, we have applied five geometric image augmentation methods that preserve the semantic information on the training images using the Albumentations library (Buslaev et al. 2020).”
MethodVehicle Damage Severity Estimation for Insurance Operations Using In-The-Wild Mobile Images
Mallios, D., Li, X., McLaughlin, N., del Rincón, J. M., Galbraith, C., & Garland, R. (2023) · IEEE Access · doi:10.1109/access.2023.3299223 · Cited by 32
Zong, Z., Chen, C., Mi, X., Sun, W., Song, Y., Li, J., Dong, Z., Huang, R., & Yang, B. (2019) · The international archives of the photogrammetry, remote sensing and spatial information sciences/International archives of the photogrammetry, remote sensing and spatial information sciences · doi:10.5194/isprs-archives-xlii-2-w16-293-2019 · Cited by 31
Automatic salt deposits segmentation: A deep learning approachPreprint
Karchevskiy, M., Ashrapov, I., & Kozinkin, L. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1812.01429 · Cited by 30
“To increase the train dataset we used some augmentation techniques [9].”
MethodMultimodal Approach for DeepFake Detection
Lomnitz, M., Hampel-Arias, Z., Sandesara, V., & Hu, S. (2020) · 2020 IEEE Applied Imagery Pattern Recognition Workshop (AIPR) · doi:10.1109/aipr50011.2020.9425192 · Cited by 30
“To increase the variance in the data we used the Albumentations [43] package for data augmentation, which included random translations, rotations, rescale, horizontal flips, brightness and contrast shifts, motion blur, gaussian noise and jpeg compression.”
MethodResearch on Real Time Obstacle Avoidance Method for AUV Based on Combination of ViT-DPFN and MPC
Zheng, H., Sun, Y., Zhang, G., Zhang, L., & Zhang, W. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3348890 · Cited by 34
Estimating hydroponic lettuce phenotypic parameters for efficient resource allocation
Ojo, M. O., Zahid, A., & Masabni, J. G. (2024) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2024.108642 · Cited by 32
ConTEXTual Net: A Multimodal Vision-Language Model for Segmentation of Pneumothorax
Huemann, Z., Tie, X., Hu, J., & Bradshaw, T. (2024) · Journal of Imaging Informatics in Medicine · doi:10.1007/s10278-024-01051-8 · Cited by 37
Segmenting medical images with limited data
Liu, Z., Lv, Q., Lee, C. H., & Shen, L. (2024) · Neural Networks · doi:10.1016/j.neunet.2024.106367 · Cited by 35
“We emphasize that the OAA can be seamlessly integrated into the established DA pipeline [59] with one line of code, and utilizing it is as straightforward as leveraging arbitrary conventional DA transformations.”
MethodLiu, W., Chen, Y., Lu, Z., Lu, X., Wu, Z. W., Zheng, Z., Suo, Y., Lan, C., & Yuan, X. (2024) · Plant Phenomics · doi:10.34133/plantphenomics.0201 · Cited by 34
“To address this issue, the following geometric transformations were randomly applied to each image in the training dataset of the leaf segmentation network and the lesion segmentation network using Albumentation [34] as the data augmentation library.”
MethodDiao, Z., Ueda, K., Hou, L., Li, F., Yamashita, H., & Abe, M. (2024) · Small Methods · doi:10.1002/smtd.202400813 · Cited by 30
“Then, data augmentation 60 , which contained random affine transform, image coping, and contrast change, was separately applied to the training dataset and the validation dataset and expands the dataset by 5 times.”
MethodSynthetic Data Augmentation for Zero-Shot Cross-Lingual Question Answering
Riabi, A., Scialom, T., Keraron, R., Sagot, B., Seddah, D., & Staiano, J. (2021) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2021.emnlp-main.562 · Cited by 67
“It has found successful application in several areas, such as time series analysis (Forestier et al., 2017) and computer vision (Buslaev et al., 2020).”
BackgroundInsect recognition based on complementary features from multiple views
An, J., Du, Y., Hong, P., Zhang, L., & Weng, X. (2023) · Scientific Reports · doi:10.1038/s41598-023-29600-1 · Cited by 35
UAV-Based Image and LiDAR Fusion for Pavement Crack Segmentation
Elamin, A., & El‐Rabbany, A. (2023) · Sensors · doi:10.3390/s23239315 · Cited by 36
“The augmentations included random rotations and flips using the Albumentations library [51].”
MethodMulti-Task Visual Perception for Object Detection and Semantic Segmentation in Intelligent Driving
Zhan, J., Liu, J., Wu, Y., & Guo, C. (2024) · Remote Sensing · doi:10.3390/rs16101774 · Cited by 32
“Albumentations [53] is a powerful image augmentation library with many different image transformation operations.”
BackgroundMulti-scale morphology-aided deep medical image segmentation
Ghosh, S., & Das, S. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.109047 · Cited by 37
Banasiak, P., Berezowski, P., Zapłata, R., Mielcarek, M., Duraj, K. M., & Stereńczak, K. (2022) · Remote Sensing · doi:10.3390/rs14040995 · Cited by 32
Le, W., Vorontsov, E., Romero, F. P., Seddik, L., Elsharief, M. M., Nguyen‐Tan, P. F., Roberge, D., Bahig, H., & Kadoury, S. (2022) · Scientific Reports · doi:10.1038/s41598-022-07034-5 · Cited by 37
Oliveira, B., Torres, H. R., Morais, P., Veloso, F., Baptista, A. L., Fonseca, J. C., & Vilaça, J. L. (2023) · Scientific Reports · doi:10.1038/s41598-022-27089-8 · Cited by 36
High-Order Semantic Decoupling Network for Remote Sensing Image Semantic Segmentation
Zheng, C., Nie, J., Wang, Z., Song, N. X., Wang, J., & Wei, Z. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3249230 · Cited by 31
“we utilize the albumentations library [39] for data augmentation and normalize all training images to [0.”
MethodShape-Aware Monocular 3D Object Detection
Chen, W., Zhao, J., Zhao, W., & Wu, S.-Y. (2023) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2023.3249909 · Cited by 32
“augmentation tool from [41] is adopted.”
A framework to estimate generating capacities of PV systems using satellite imagery segmentation
Jurakuziev, D., Jumaboev, S., & Lee, M. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.106186 · Cited by 32
Semi-Supervised Modified-UNet for Lung Infection Image Segmentation
Upadhyay, A. K., & Bhandari, A. K. (2023) · IEEE Transactions on Radiation and Plasma Medical Sciences · doi:10.1109/trpms.2023.3272209 · Cited by 29
“[71] for advanced levels of image augmentations.”
Belcher, B. T., Bower, E. H., Burford, B. P., Celis, M. R., Fahimipour, A. K., Guevara, I., Katija, K., Khokhar, Z., Manjunath, A., Nelson, S., et al. (2023) · Frontiers in Marine Science · doi:10.3389/fmars.2023.1157370 · Cited by 36
MedAugment: Universal automatic data augmentation plug-in for medical image analysis
Liu, Z., Lv, Q., Li, Y., Yang, Z., & Shen, L. (2026) · Knowledge-Based Systems · doi:10.1016/j.knosys.2026.115828 · Cited by 43
Fdez-Vidal, X. R., Fernández‐Canto, N., Romero‐Rodríguez, M. Á., Ramos‐Cabrer, A. M., Pereira‐Lorenzo, S., & Lombardero, M. (2023) · Food Control · doi:10.1016/j.foodcont.2023.110198 · Cited by 29
Lung pneumonia severity scoring in chest X-ray images using transformers
Slika, B., Dornaika, F., Merdji, H., & Hammoudi, K. (2024) · Medical & Biological Engineering & Computing · doi:10.1007/s11517-024-03066-3 · Cited by 32
“Horizontal image flipping and image blurring are two traditional augmentation methods that were also tested [65].”
MethodNguyen, T. T. H., Nguyen, L. P. T., & Cao, H. (2024) · Information Fusion · doi:10.1016/j.inffus.2024.102782 · Cited by 47
Integration of Deep Learning Vision Systems in Collaborative Robotics for Real-Time Applications
Terras, N., Pereira, F. A. D. S., Silva, A. R., Santos, A. A., Lopes, A. M., da Silva, A. F., Cartal, L. A., Apostolescu, T. C., Badea, F., & Machado, J. (2025) · Applied Sciences · doi:10.3390/app15031336 · Cited by 32
Illarionova, S., Shadrin, D., Gubanov, F., Shutov, M., Tasuev, U., Evteeva, K., Mironenko, M., & Burnaev, E. V. (2025) · Scientific Reports · doi:10.1038/s41598-025-94002-4 · Cited by 32
“We also used data augmentation techniques, such as rotation and elastic transformations 63 , to further enhance the diversity of our sample set.”
Dental caries detection in children using intraoral scans and deep learning
Jones, B., Lambach, M. S., Chen, T., Michou, S., Kilpatrick, N. M., Curtis, N. C., Burgner, D. P., Vannahme, C., & Silva, M. (2025) · Journal of Dentistry · doi:10.1016/j.jdent.2025.105906 · Cited by 33
Culture-free detection of bacteria from blood for rapid sepsis diagnosis
Miguélez, M. H. M., Osaid, M., Hallström, E., Kaya, K., Larsson, J., Kandavalli, V. K., Wählby, C., Elf, J., & van der Wijngaart, W. (2025) · npj Digital Medicine · doi:10.1038/s41746-025-01948-w · Cited by 31
Improving deep learning sorghum head detection through test time augmentation
Gonzalo‐Martín, C., Garcia-Pedrero, A. M., & Lillo‐Saavedra, M. (2021) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2021.106179 · Cited by 31
Feng, Y., Liu, S., Cheng, Z., Quiroz, J. C., Rezazadegan, D., Chen, P.-K., Lin, Q., Qian, L., Liu, X., Berkovsky, S., et al. (2021) · Information · doi:10.3390/info12110471 · Cited by 28
Siciarz, P., & McCurdy, B. M. C. (2022) · Physics in Medicine and Biology · doi:10.1088/1361-6560/ac530e · Cited by 28
Using Neural Encoder-Decoder Models With Continuous Outputs for Remote Sensing Image Captioning
Ramos, R., & Martins, B. (2022) · IEEE Access · doi:10.1109/access.2022.3151874 · Cited by 33
Gafurov, A. N., Phung, T. H., Ryu, B.-H., Kim, I., & Lee, T. (2022) · International Journal of Precision Engineering and Manufacturing-Green Technology · doi:10.1007/s40684-022-00461-9 · Cited by 28
“To prevent overfitting during training, the images and label images were augmented using the following effects: horizontal and vertical flip, rotation, brightness and contrast adjustments, and Gaussian blur [32].”
MethodMendel, R., Rauber, D., de Souza, L. A., Jr., Papa, J. P., & Palm, C. (2023) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2023.106585 · Cited by 31
Deep learning-enabled segmentation of ambiguous bioimages with deepflash2
Griebel, M., Segebarth, D., Stein, N., Schukraft, N., Tovote, P., Blum, R., & Flath, C. M. (2023) · Nature Communications · doi:10.1038/s41467-023-36960-9 · Cited by 33
“The DL-related part is built upon the rich ecosystem of PyTorch 34 libraries, comprising fastai 35 for thetrainingprocedure, segmentationmodelspytorch 28 forsegmentation architectures, timm 30 for pretrained encoders, and albumentations 36 for data augmentations.”
MethodZhu, J., Zhao, D., & Luo, X. (2023) · Road Materials and Pavement Design · doi:10.1080/14680629.2023.2199880 · Cited by 30
“According to previous studies, the optimal expansion of the dataset was determined to be 30% (Buslaev et al., 2020; Cubuk et al., 2019).”
MethodJensen, A., Flood, P. D. L., Palm-Vlasak, L. S., Burton, W. S., Chevalier, A., Rullkoetter, P. J., & Banks, S. A. (2023) · The Journal of Arthroplasty · doi:10.1016/j.arth.2023.05.029 · Cited by 28
Land Use and Land Cover Mapping in China Using Multimodal Fine-Grained Dual Network
Liu, S., Wang, H., Hu, Y., Zhang, M., Zhu, Y., Wang, Z., Li, D., Yang, M., & Wang, F. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3285912 · Cited by 29
“During training, we use augmentation methods such as random resize with interpolation rates between 0.5 and 1.5, random crop with size 768 × 768, and gridshuffle [64], which grid number is set to 4.”
MethodInfluentialSupervised fine-tuned approach for automated detection of diabetic retinopathy
Ohri, K., & Kumar, M. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-16049-7 · Cited by 33
“For this, albumen-tations library is used which is a fast and flexible image augmentation library used in deep learning and computer vision tasks [7].”
Weighted Ensemble Object Detection with Optimized Coefficients for Remote Sensing Images
Körez, A., Barışçı, N., Çetin, A., & Ergün, U. (2020) · ISPRS International Journal of Geo-Information · doi:10.3390/ijgi9060370 · Cited by 28
Liu, Q., Kampffmeyer, M., Jenssen, R., & Salberg, A.-B. (2021) · International Journal of Remote Sensing · doi:10.1080/01431161.2021.1936267 · Cited by 34
“The albumentations library (Buslaev et al. 2020) for data augmentation is utilized in this work.”
MethodVideo Face Manipulation Detection Through Ensemble of CNNs
Bonettini, N., Cannas, E. D., Mandelli, S., Bondi, L., Bestagini, P., & Tubaro, S. (2021) · 2020 25th International Conference on Pattern Recognition (ICPR) · doi:10.1109/icpr48806.2021.9412711 · Cited by 27
“Specifically, we resort to Albumentation [36] as our data-augmentation library, while we use Pytorch [37] as Deep Learning framework.”
MethodDeep-learning model for evaluating histopathology of acute renal tubular injury
Nguyen, T. T. U., Nguyen, A.-T., Kim, H., Jung, Y. J., Park, W. S., Kim, K. M., Park, I., & Kim, W. B. (2024) · Scientific Reports · doi:10.1038/s41598-024-58506-9 · Cited by 29
Coarse-to-fine semantic segmentation of satellite images
Chen, H., Yang, W. T., Liu, L., & Xia, G.-S. (2024) · ISPRS Journal of Photogrammetry and Remote Sensing · doi:10.1016/j.isprsjprs.2024.07.028 · Cited by 33
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.