Few-Shot Image Classification via Contrastive Self-Supervised Learning
This addresses the need for more flexible and data-efficient few-shot learning methods in computer vision, though it is incremental as it builds on existing self-supervised and few-shot techniques.
The paper tackles the problem of few-shot image classification by introducing an unsupervised approach that uses contrastive self-supervised learning for meta-training and graph aggregation for classifier training, achieving state-of-the-art performance with an 8-28% increase over existing unsupervised methods.
Most previous few-shot learning algorithms are based on meta-training with fake few-shot tasks as training samples, where large labeled base classes are required. The trained model is also limited by the type of tasks. In this paper we propose a new paradigm of unsupervised few-shot learning to repair the deficiencies. We solve the few-shot tasks in two phases: meta-training a transferable feature extractor via contrastive self-supervised learning and training a classifier using graph aggregation, self-distillation and manifold augmentation. Once meta-trained, the model can be used in any type of tasks with a task-dependent classifier training. Our method achieves state of-the-art performance in a variety of established few-shot tasks on the standard few-shot visual classification datasets, with an 8- 28% increase compared to the available unsupervised few-shot learning methods.