An efficient Transformer with neighborhood contrastive tokenization for hyperspectral images classification

The success of vision Transformers (ViTs) relies heavily on the self-attention mechanism, which requires support from appropriate patch tokenization. However, hyperspectral image (HSI) often suffer from significant noise distortions and spectral uncertainty, which result in unstable attention patter...

Descripció completa

Dades bibliogràfiques

Autors principals:	Miaomiao Liang, Xianhao Zhang, Xiangchun Yu, Lingjuan Yu, Zhe Meng, Xiaohong Zhang, Licheng Jiao
Format:	Artigo
Idioma:	Inglês
Publicat:	Elsevier 2024-07-01
Col·lecció:	International Journal of Applied Earth Observations and Geoinformation
Matèries:	Hyperspectral image (HSI) Transformer Contrastive tokenization Self-supervised learning (SSL)
Accés en línia:	http://www.sciencedirect.com/science/article/pii/S1569843224003339
Etiquetes:	Afegir etiqueta Sense etiquetes, Sigues el primer a etiquetar aquest registre!