Enhancing Few-Shot Image Classification With a Multi-Faceted Self-Supervised and Contrastive Learning Approach

One effective approach for solving few-shot classification is learning deep representations that measure the similarity between query images and a few support images of specific categories. Recent methods overly relied on a single metric, resulting in insufficient interdependence in image feature re...

ver descrição completa

Detalhes bibliográficos

Principais autores:	Ling Hu, Wei Wu
Formato:	Artigo
Idioma:	Inglês
Publicado em:	IEEE 2024-01-01
coleção:	IEEE Access
Assuntos:	Few-shot learning feature reconstruction self-supervised learning contrastive learning Brownian distance covariance
Acesso em linha:	https://ieeexplore.ieee.org/document/10746489/
Tags:	Adicionar Tag Sem tags, seja o primeiro a adicionar uma tag!