论视觉知识(英文)

被引:22
作者
Yun-he PAN
机构
[1] ZhejiangUniversity
关键词
AI; 知识表达;
D O I
暂无
中图分类号
TP18 [人工智能理论]; TP391.41 [];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ; 080203 ;
摘要
提出"视觉知识"概念.视觉知识是知识表达的一种新形式.它与迄今为止人工智能(AI)所用知识表达方法不同.其中视觉概念具有典型(prototype)与范畴结构、层次结构与动作结构等要素.视觉概念能构成视觉命题,包括场景结构与动态结构,视觉命题能构成视觉叙事.指出重构计算机图形学成果可实现视觉知识表达及其推理与操作,重构计算机视觉成果可实现视觉知识学习.实现视觉知识表达、推理、学习和应用技术将是AI 2.0取得突破的重要方向之一。
引用
收藏
页码:1021 / 1026
页数:6
相关论文
共 11 条
[1]  
Computer Vision:Fundamentals of the Computational Theory and Algorithms. Ma SD,Zhang ZY. China Science Publishing&Media Ltd . 1998
[2]  
Computer Graphics:Principles,Methods and Applications. Pan YH,Tong RF,Tang M. Higher Education Press . 2011
[3]  
Cognitive Psychology. Anderson JR,Yang Q,Trans. Jilin Education Press,China . 1989
[4]  
A study on the imagery information model of imagery thinking. Pan YH. Patt Recogn Artif Intell . 1991
[5]  
Stacked capsule autoencoders. Kosiorek AR,Sabour S,Teh YW,et al. https://arxiv.org/abs/190606818 . 2019
[6]  
Action and object words are differentially anchored in the sensory motor system-a perspective on cognitive embodiment. Horoufchin H,Bzdok D,Buccino G,et al. Sci Reports . 2018
[7]  
Multi-object representation learning with iterative variational inference. Greff K,Kaufmann RL,Kabra R,et al. https://arxiv.org/abs/1903.00450 . 2019
[8]  
50 years of test(un)fairness:lessons for machine learning. Hutchinson B,Mitchell M. Proc Conf on Fairness,Accountability,and Transparency . 2019
[9]  
Heading toward Artificial Intelligence 2.0[J] . Yunhe Pan. &nbspEngineering . 2016 (4)
[10]  
3D point capsule networks. Zhao Y,Birdal T,Deng H,et al. Proc IEEE Conf on Computer Vision and Pattern Recognition . 2019