Current state-of-the-art approaches for few-shot action recognition achieve promising
performance by conducting frame-level matching on learned visual features. However, they …

Teaching machines to recognize a new category based on few training samples especially
only one remains challenging owing to the incomprehensive understanding of the novel …

We are interested in develo** a unified machine learning framework for effectively training
machine learning models from many small data sources such as mobile devices. This is a …

This paper focuses on skeleton-based few-shot action recognition. Since skeleton is
essentially a sparse representation of human action, the feature maps extracted from it …

Humans exhibit a remarkable ability to learn quickly from a limited number of labeled
samples, a capability that starkly contrasts with that of current machine learning systems …

Metric-based few-shot learning categorizes unseen query instances by measuring their
distance to the categories appearing in the given support set. To facilitate distance …

Transfer learning has been widely adopted for few-shot classification. Recent studies reveal
that obtaining good generalization representation of images on novel classes is the key to …

Recently, transfer learning has generated promising performance in few-shot classification
by pre-training a backbone network on base classes and then applying it to novel classes …

Fine-grained action recognition is attracting increasing attention due to the emerging
demand of specific action understanding in real-world applications, whereas the data of rare …

Gesture recognition and its applications have been widely studied and received much
attention in recent years. Existing works on hand gesture recognition aim to train …