Cargando…

Deep learning models fail to capture the configural nature of human shape perception

A hallmark of human object perception is sensitivity to the holistic configuration of the local shape features of an object. Deep convolutional neural networks (DCNNs) are currently the dominant models for object recognition processing in the visual cortex, but do they capture this configural sensit...

Descripción completa

Detalles Bibliográficos
Autores principales: Baker, Nicholas, Elder, James H.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Elsevier 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9429800/
https://www.ncbi.nlm.nih.gov/pubmed/36060067
http://dx.doi.org/10.1016/j.isci.2022.104913