基于28nm二维/三维统一稀疏卷积加速器的大规模点云网络演示平台

2023 IEEE 5th International Conference on Artificial Intelligence Circuits and Systems (AICAS) Pub Date : 2023-06-11 DOI:10.1109/AICAS57966.2023.10168558

Xiaoyu Feng, Wenyu Sun, Shupei Fan, Chen Tang, Yixiong Yang, Jinshan Yue, Q. Liao, Huazhong Yang, Yongpan Liu

{"title":"基于28nm二维/三维统一稀疏卷积加速器的大规模点云网络演示平台","authors":"Xiaoyu Feng, Wenyu Sun, Shupei Fan, Chen Tang, Yixiong Yang, Jinshan Yue, Q. Liao, Huazhong Yang, Yongpan Liu","doi":"10.1109/AICAS57966.2023.10168558","DOIUrl":null,"url":null,"abstract":"3D point cloud processing plays an important role in many emerging applications such as autonomous driving, visual navigation, and virtual reality. It calls for hardware acceleration of multiple key operations, including 3D Submanifold SCONV, 3D non-Submanifold SCONV, and 2D SCONV. This work presents a 2D/3D unified sparse convolution accelerator for large-scale voxel-based point cloud networks. The chip is fabricated in TSMC 28nm CMOS technology to achieve 3.3-16.9 FPS running from 60-400MHz when computing the SECOND network on KITTI dataset. This work has been included by ISSCC2023 [1]. A demonstration is given to show the real-time 3D processing with a lidar sensor.","PeriodicalId":296649,"journal":{"name":"2023 IEEE 5th International Conference on Artificial Intelligence Circuits and Systems (AICAS)","volume":"107 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"A Demonstration Platform for Large-Scaled Point Cloud Network Based on 28nm 2D/3D Unified Sparse Convolution Accelerator\",\"authors\":\"Xiaoyu Feng, Wenyu Sun, Shupei Fan, Chen Tang, Yixiong Yang, Jinshan Yue, Q. Liao, Huazhong Yang, Yongpan Liu\",\"doi\":\"10.1109/AICAS57966.2023.10168558\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"3D point cloud processing plays an important role in many emerging applications such as autonomous driving, visual navigation, and virtual reality. It calls for hardware acceleration of multiple key operations, including 3D Submanifold SCONV, 3D non-Submanifold SCONV, and 2D SCONV. This work presents a 2D/3D unified sparse convolution accelerator for large-scale voxel-based point cloud networks. The chip is fabricated in TSMC 28nm CMOS technology to achieve 3.3-16.9 FPS running from 60-400MHz when computing the SECOND network on KITTI dataset. This work has been included by ISSCC2023 [1]. A demonstration is given to show the real-time 3D processing with a lidar sensor.\",\"PeriodicalId\":296649,\"journal\":{\"name\":\"2023 IEEE 5th International Conference on Artificial Intelligence Circuits and Systems (AICAS)\",\"volume\":\"107 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-06-11\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2023 IEEE 5th International Conference on Artificial Intelligence Circuits and Systems (AICAS)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/AICAS57966.2023.10168558\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2023 IEEE 5th International Conference on Artificial Intelligence Circuits and Systems (AICAS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/AICAS57966.2023.10168558","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

摘要

三维点云处理在自动驾驶、视觉导航、虚拟现实等新兴应用中发挥着重要作用。它需要多个关键操作的硬件加速，包括3D Submanifold SCONV、3D non-Submanifold SCONV和2D SCONV。本文提出了一种用于大规模体素点云网络的2D/3D统一稀疏卷积加速器。该芯片采用台积电28nm CMOS工艺制造，在KITTI数据集上计算SECOND网络时，在60-400MHz范围内实现3.3-16.9 FPS。该工作已被ISSCC2023收录[1]。最后给出了用激光雷达传感器进行实时三维处理的实例。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

A Demonstration Platform for Large-Scaled Point Cloud Network Based on 28nm 2D/3D Unified Sparse Convolution Accelerator

3D point cloud processing plays an important role in many emerging applications such as autonomous driving, visual navigation, and virtual reality. It calls for hardware acceleration of multiple key operations, including 3D Submanifold SCONV, 3D non-Submanifold SCONV, and 2D SCONV. This work presents a 2D/3D unified sparse convolution accelerator for large-scale voxel-based point cloud networks. The chip is fabricated in TSMC 28nm CMOS technology to achieve 3.3-16.9 FPS running from 60-400MHz when computing the SECOND network on KITTI dataset. This work has been included by ISSCC2023 [1]. A demonstration is given to show the real-time 3D processing with a lidar sensor.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

2023 IEEE 5th International Conference on Artificial Intelligence Circuits and Systems (AICAS)

自引率

0.00%

发文量