From Sample Poverty to Rich Feature Learning: A New Metric Learning Method for Few-Shot Classification
With the rapid development of deep learning in the field of computer vision, few-shot learning has emerged as an effective approach to tackle the challenge of data scarcity, garnering widespread attention from researchers.Despite significant progress in few-shot learning, current few-shot image classification methods have not fully exploited the fe