Corpus ID: 211532796

Understanding and Enhancing Mixed Sample Data Augmentation

  title={Understanding and Enhancing Mixed Sample Data Augmentation},
  author={E. Harris and A. Marcu and Matthew Painter and M. Niranjan and A. Pr{\"u}gel-Bennett and J. Hare},
  • E. Harris, A. Marcu, +3 authors J. Hare
  • Published 2020
  • Computer Science
  • ArXiv
  • Mixed Sample Data Augmentation (MSDA) has received increasing attention in recent years, with many successful variants such as MixUp and CutMix. Following insight on the efficacy of CutMix in particular, we propose FMix, an MSDA that uses binary masks obtained by applying a threshold to low frequency images sampled from Fourier space. FMix improves performance over MixUp and CutMix for a number of state-of-theart models across a range of data sets and problem settings. We go on to analyse MixUp… CONTINUE READING
    8 Citations

    Figures, Tables, and Topics from this paper

    Explore Further: Topics Discussed in This Paper

    Grid cut and mix: flexible and efficient data augmentation
    • Highly Influenced
    An Empirical Analysis of the Impact of Data Augmentation on Knowledge Distillation
    • 1
    • Highly Influenced
    • PDF
    Milking CowMask for Semi-Supervised Image Classification
    • 4
    • PDF
    Causal Explanations of Image Misclassifications
    • PDF
    SpinalNet: Deep Neural Network with Gradual Input
    • 5
    • PDF


    Improved Mixed-Example Data Augmentation
    • 34
    • PDF
    MixUp as Locally Linear Out-Of-Manifold Regularization
    • 71
    • PDF
    Understanding Mixup Training Methods
    • 16
    • Highly Influential
    Data Augmentation Revisited: Rethinking the Distribution Gap between Clean and Augmented Data
    • 15
    • PDF
    Improved Regularization of Convolutional Neural Networks with Cutout
    • 852
    • PDF
    Data Augmentation by Pairing Samples for Images Classification
    • H. Inoue
    • Mathematics, Computer Science
    • ArXiv
    • 2018
    • 138
    • PDF
    Data Augmentation Using Random Image Cropping and Patching for Deep CNNs
    • 49
    • PDF
    CutMix: Regularization Strategy to Train Strong Classifiers With Localizable Features
    • 312
    • PDF
    mixup: Beyond Empirical Risk Minimization
    • 1,378
    • Highly Influential
    • PDF
    Deep Pyramidal Residual Networks
    • 291
    • PDF