⚡ 本页包含 AI 生成的分析内容，仅供参考
该论文提出了一种基于新皮层计算模型的36核视觉识别处理器，采用Kautz片上网络实现2.3Tb/s的高带宽通信，在1.0TOPS/W的能效下支持通用视觉识别任务，解决了传统异构多核架构只能处理预定义窄任务的问题。
Unlike human brains, where various kinds of visual recognition tasks are carried out with homogeneous neocortical circuits and a unified working mechanism, existing visual recognition processors [1-4] rely on multiple algorithms and heterogeneous multicore architectures to accomplish their narrowly predefined recognition tasks. In contrast, new brain-mimicking recognition algorithms have been proposed recently [5,6]; we label such algorithms as belonging to the Neocortical Computing (NC) model. Based on neuroscience findings, NC algorithms model both the human brain’s static and dynamic visual recognition streams (as shown in Fig. 28.2.1) through a series of unified matching/pooling operations. The algorithms exhibit high visual recognition capability and perform accurately on wide range of image/video recognition tasks. However, using the NC model for real-time recognition involves several hundred GOPS of both
Chuan-Yung Tsai, Yu-Ju Lee, Chun-Ting Chen, Liang-Gee Chen
National Taiwan University, Taipei, Taiwan