首页> 外文会议>International Conference on Neural Information Processing >Towards a Visual Microprocessor with TeraOPS Speed and Kilo Real-time Frame Rate
【24h】

Towards a Visual Microprocessor with TeraOPS Speed and Kilo Real-time Frame Rate

机译:朝着具有TeraOps Speed和Kilo实时帧速率的视觉微处理器

获取原文

摘要

The analogic (analog and logic) cellular computer based on the CNN Universal Machine (CNN-UM) architecture is presented as a stored program visual microprocessor. Its CMOS mixed-signal implementation with focal-plane optical sensors provides for a new quality in single-chip computers with TeraOps speed and Kilo real-time frame rate. After a short introduction to the Cellular Neural/Nonlinear Network (CNN) and the CNN-UM the computational infrastructure and some application case studies are presented.
机译:基于CNN通用机器(CNN-UM)架构的模拟(模拟和逻辑)蜂窝计算机被呈现为存储的程序视觉微处理器。其CMOS混合信号实现,具有焦面光学传感器,在具有TeraOps Speed和Kilo实时帧速率的单片机中提供了新的质量。在蜂窝神经/非线性网络(CNN)和CNN-UM的简要介绍之后,提出了计算基础设施和一些应用案例研究。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号