Class-relevant Patch Embedding Selection for Few-Shot Image Classification

arXiv:2405.03722v1 Announce Type: new
Abstract: Effective image classification hinges on discerning relevant features from both foreground and background elements, with the foreground typically holding the critical information. While humans adeptly classify images with limited exposure, artificial neural networks often struggle with feature selection from rare samples. To address this challenge, we propose a novel method for selecting class-relevant patch embeddings. Our approach involves splitting support and query images into patches, encoding them using a pre-trained Vision Transformer (ViT) to obtain …

abstract artificial artificial neural networks arxiv challenge class classification cs.cv embedding feature features feature selection few-shot humans image images information networks neural networks samples struggle type while