Frozen Feature Augmentation for Few-Shot Image Classification
This work addresses the challenge of enhancing few-shot learning efficiency for computer vision applications, though it is incremental as it adapts standard data augmentation techniques to a frozen feature setting.
The paper tackles the problem of improving few-shot image classification by applying data augmentations directly in the frozen feature space, showing that simple augmentations like brightness consistently boost performance across multiple architectures and datasets.
Training a linear classifier or lightweight model on top of pretrained vision model outputs, so-called 'frozen features', leads to impressive performance on a number of downstream few-shot tasks. Currently, frozen features are not modified during training. On the other hand, when networks are trained directly on images, data augmentation is a standard recipe that improves performance with no substantial overhead. In this paper, we conduct an extensive pilot study on few-shot image classification that explores applying data augmentations in the frozen feature space, dubbed 'frozen feature augmentation (FroFA)', covering twenty augmentations in total. Our study demonstrates that adopting a deceptively simple pointwise FroFA, such as brightness, can improve few-shot performance consistently across three network architectures, three large pretraining datasets, and eight transfer datasets.