Augmentor: An Image Augmentation Library for Machine Learning

By Marcus D. Bloice, Christof Stocker, Andreas Holzinger. 2017

293 citations recorded by the source corpus.

The generation of artificial data based on existing observations, known as data augmentation, is a technique used in machine learning to improve model accuracy, generalisation, and to control overfitting. Augmentor is a software package, available in both Python and Julia versions, that provides a high level API for the expansion of image data using a stochastic, pipeline-based approach which effectively allows for images to be sampled from a distribution of augmented images at runtime. Augmentor provides methods for most standard augmentation practices as well as several advanced features such as label-preserving, randomised elastic distortions, and provides many helper functions for typical augmentation tasks used in machine learning.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, ML Systems & Hardware

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Antreas Antoniou
2. Amos J. Storkey
3. Heimo Müller
4. Harrison Edwards
5. Kurt Zatloukal
6. Bernd Malle
7. Andreas Holzinger
8. Robert Reihs

In Computer Vision and Pattern Recognition

1. Antreas Antoniou
2. Amos J. Storkey
3. Dacheng Tao
4. Luc Van Gool
5. Harrison Edwards
6. Bernt Schiele
7. Andreas Holzinger
8. Xiaogang Wang

In ML Systems & Hardware

1. Amos J. Storkey
2. Andreas Holzinger

Citations in the corpus

• Augmenting Image Classifiers Using Data Augmentation Generative Adversarial Networks (2018, background): “Additional papers that attempt to learn models for data augmentation include [9, 4, 30].”
• CLoDSA: a tool for augmentation in classification, localization, detection, semantic segmentation and instance segmentation tasks (2019, methodology): “Due to this fact, several libraries, like Augmentor [10] or Imgaug [11], and deep learning frameworks, like Keras [12] or Tensor-flow [13], provide features for data augmentation in the context of object classification.”
• Towards the Augmented Pathologist: Challenges of Explainable-AI in Digital Pathology (2017, background): “In our domain image augmentation (Bloice et al., 2017) can be particularly beneficial, where additional training samples are generated via variation of the given data, e.g. via rotation, elastic deformation, adding noise, etc.”