PNNPU: A 11.9 TOPS/W High-speed 3D Point Cloud-based Neural Network Processor with Block-based Point Processing for Regular DRAM Access

Sangjin Kim, Juhyoung Lee, Dongseok Im, H. Yoo
{"title":"PNNPU: A 11.9 TOPS/W High-speed 3D Point Cloud-based Neural Network Processor with Block-based Point Processing for Regular DRAM Access","authors":"Sangjin Kim, Juhyoung Lee, Dongseok Im, H. Yoo","doi":"10.23919/VLSICircuits52068.2021.9492450","DOIUrl":null,"url":null,"abstract":"An efficient and high-speed 3D point cloud-based neural network processing unit (PNNPU) is proposed using the block-based point processing. It has three key features: 1) page-based point block memory management unit (PMMU) with linked list-based page table (LLPT) for on-chip memory footprint reduction, 2) hierarchical block-wise farthest point sampling (HFPS), and block skipping ball-query (BSBQ) for fast and efficient point processing, 3) Skipping-based max-pooling prediction (SMPP) for throughput enhancement. The PNNPU is fabricated in 65nm CMOS process and evaluated on the 3D object detection (3D OD) application. As a result, it shows 84.8 fps at 266.8mW power consumption and achieving 6.6-11.9 TOPS/W energy efficiency.","PeriodicalId":106356,"journal":{"name":"2021 Symposium on VLSI Circuits","volume":"66 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-06-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"8","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 Symposium on VLSI Circuits","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.23919/VLSICircuits52068.2021.9492450","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 8

Abstract

An efficient and high-speed 3D point cloud-based neural network processing unit (PNNPU) is proposed using the block-based point processing. It has three key features: 1) page-based point block memory management unit (PMMU) with linked list-based page table (LLPT) for on-chip memory footprint reduction, 2) hierarchical block-wise farthest point sampling (HFPS), and block skipping ball-query (BSBQ) for fast and efficient point processing, 3) Skipping-based max-pooling prediction (SMPP) for throughput enhancement. The PNNPU is fabricated in 65nm CMOS process and evaluated on the 3D object detection (3D OD) application. As a result, it shows 84.8 fps at 266.8mW power consumption and achieving 6.6-11.9 TOPS/W energy efficiency.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
PNNPU:一个11.9 TOPS/W的高速3D点云神经网络处理器,具有基于块的点处理,用于常规DRAM访问
提出了一种基于分块的三维点云神经网络处理单元(PNNPU)。它有三个关键特性:1)基于页的点块内存管理单元(PMMU)和基于链表的页表(LLPT),用于减少片上内存占用;2)分层块方向最远点采样(HFPS)和块跳球查询(BSBQ),用于快速高效的点处理;3)基于跳的最大池预测(SMPP),用于提高吞吐量。采用65nm CMOS工艺制备了PNNPU,并对其3D目标检测(3D OD)应用进行了评估。因此,它在266.8mW的功耗下显示了84.8 fps,实现了6.6-11.9 TOPS/W的能效。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
PIMCA: A 3.4-Mb Programmable In-Memory Computing Accelerator in 28nm for On-Chip DNN Inference A 24–31 GHz Reference Oversampling ADPLL Achieving FoMjitter−N of -269.3 dB A 6.78 MHz Wireless Power Transfer System for Simultaneous Charging of Multiple Receivers with Maximum Efficiency using Adaptive Magnetic Field Distributor IC Enhanced Core Circuits for scaling DRAM: 0.7V VCC with Long Retention 138ms at 125°C and Random Row/Column Access Times Accelerated by 1.5ns A Sub-mW Dual-Engine ML Inference System-on-Chip for Complete End-to-End Face-Analysis at the Edge
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1