ChimeraMix: Image Classification on Small Datasets via Masked Feature Mixing
This addresses the limitation of requiring large labeled datasets for real-world applications, offering a solution for small-data scenarios.
The paper tackles the problem of training deep neural networks on small datasets by proposing ChimeraMix, a method that generates new samples through masked feature mixing, achieving superior performance on benchmark datasets like ciFAIR-10, STL-10, and ciFAIR-100.
Deep convolutional neural networks require large amounts of labeled data samples. For many real-world applications, this is a major limitation which is commonly treated by augmentation methods. In this work, we address the problem of learning deep neural networks on small datasets. Our proposed architecture called ChimeraMix learns a data augmentation by generating compositions of instances. The generative model encodes images in pairs, combines the features guided by a mask, and creates new samples. For evaluation, all methods are trained from scratch without any additional data. Several experiments on benchmark datasets, e.g. ciFAIR-10, STL-10, and ciFAIR-100, demonstrate the superior performance of ChimeraMix compared to current state-of-the-art methods for classification on small datasets.