Cargando…
Learning to see stuff
Materials with complex appearances, like textiles and foodstuffs, pose challenges for conventional theories of vision. But recent advances in unsupervised deep learning provide a framework for explaining how we learn to see them. We suggest that perception does not involve estimating physical quanti...
| Autores principales: | , |
|---|---|
| Formato: | Online Artículo Texto |
| Lenguaje: | English |
| Publicado: |
Elsevier B. V
2019
|
| Materias: | |
| Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6919301/ https://www.ncbi.nlm.nih.gov/pubmed/31886321 http://dx.doi.org/10.1016/j.cobeha.2019.07.004 |