R. Pirrone, G. Careri, F. S. Fabiano, A. Gentile, S. Gaglio
{"title":"Real-time low level feature extraction for on-board robot vision systems","authors":"R. Pirrone, G. Careri, F. S. Fabiano, A. Gentile, S. Gaglio","doi":"10.1109/CAMP.2005.44","DOIUrl":null,"url":null,"abstract":"Robot vision systems notoriously require large computing capabilities, rarely available on physical devices. Robots have limited embedded hardware, and almost all sensory computation is delegated to remote machines. Emerging gigascale integration technologies offer the opportunity to explore alternative computing architectures that can deliver a significant boost to on-board computing when implemented in embedded, reconfigurable devices. This paper explores the mapping of low level feature extraction on one such architecture, the Georgia Tech SIMD Pixel Processor (SIMPil). The Fast Boundary Web Extraction (fBWE) algorithm is adapted and mapped on SIMPil as a fixed-point, data parallel implementation. Application components and their mapping details are provided in this contribution along with a detailed analysis of their performance.","PeriodicalId":393875,"journal":{"name":"Seventh International Workshop on Computer Architecture for Machine Perception (CAMP'05)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2005-07-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Seventh International Workshop on Computer Architecture for Machine Perception (CAMP'05)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CAMP.2005.44","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Robot vision systems notoriously require large computing capabilities, rarely available on physical devices. Robots have limited embedded hardware, and almost all sensory computation is delegated to remote machines. Emerging gigascale integration technologies offer the opportunity to explore alternative computing architectures that can deliver a significant boost to on-board computing when implemented in embedded, reconfigurable devices. This paper explores the mapping of low level feature extraction on one such architecture, the Georgia Tech SIMD Pixel Processor (SIMPil). The Fast Boundary Web Extraction (fBWE) algorithm is adapted and mapped on SIMPil as a fixed-point, data parallel implementation. Application components and their mapping details are provided in this contribution along with a detailed analysis of their performance.
众所周知,机器人视觉系统需要大量的计算能力,很少在物理设备上可用。机器人的嵌入式硬件有限,几乎所有的感官计算都委托给远程机器。新兴的千兆级集成技术提供了探索替代计算架构的机会,当在嵌入式可重构设备中实现时,这些架构可以显著提升板载计算。本文探讨了低级特征提取在这样一个架构上的映射,佐治亚理工学院SIMD像素处理器(SIMPil)。将快速边界Web提取(Fast Boundary Web Extraction, fBWE)算法映射到SIMPil上,作为一个定点数据并行实现。本文提供了应用程序组件及其映射细节,以及对其性能的详细分析。