.png){kind=icon}
返回
Learning to see stuff
.png){kind=icon}
OA
AI
DOI：10.1016/j.cobeha.2019.07.004.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
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 quantities like reflectance or lighting. Instead, representations emerge from learning to encode and predict the visual input as efficiently and accurately as possible. Neural networks can be trained to compress natural images or to predict frames in movies without 'ground truth' data about the outside world. Yet, to succeed, such systems may automatically discover how to disentangle distal causal factors. Such 'statistical appearance models' potentially provide a coherent explanation of both failures and successes in perception.
Keyword:
VISUAL-PERCEPTION
NEURAL ACTIVITY
BRAIN
REPRESENTATION
OBJECTS
VISION
COLOR
MOTOR
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
3.5
论文数:
1.3K
被引数:
6.6K
机构
引用论文
Brain waves and brain wiring: The role of endogenous and sensory-driven neural activity in development.png)
PEDIATRIC RESEARCH
OAAI
PEDIATRIC RESEARCHIF3.1
OAAI分享收藏分享收藏Study of bi-directional buck-boost converter topologies for application in electrical vehicle motor drives应用于电动汽车电机驱动的双向buck-boost变换器拓扑研究
PREAI
分享收藏The response of rectangular piezoelectric sensors to Rayleigh and Lamb ultrasonic waves矩形压电传感器对瑞利和兰姆超声波的响应OAAI
OAAI分享收藏
查看更多内容