There are no files associated with this item.
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.citation.title | IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS | - |
dc.contributor.author | Jung, Jueun | - |
dc.contributor.author | Kim, Seungbin | - |
dc.contributor.author | Jang, Wuyoung | - |
dc.contributor.author | Seo, Bokyoung | - |
dc.contributor.author | Lee, Kyuho Jason | - |
dc.date.accessioned | 2024-03-14T16:05:11Z | - |
dc.date.available | 2024-03-14T16:05:11Z | - |
dc.date.created | 2024-01-08 | - |
dc.date.issued | 2024-01 | - |
dc.description.abstract | An energy-efficient, unified convolutional neural network (CNN) accelerator is proposed with a lightweight RGB-D network to achieve real-time, multi-object semantic segmentation in autonomous electric vehicle system. First, a lightweight Depth-fused Trilateral Network (DTN) is proposed to achieve high accuracy and real-time operation for road and multi-object segmentation at the same time. Optimized with various types of convolution layers and limited hardware resources, the DTN achieves 94.73% accuracy on KITTI Road dataset. Second, the unified CNN processor is designed with dual-mode shift-register-based input reconfiguration units and layer fusion architecture with 2-types of processing elements for depth-wise separable convolution (DSC) to support 5 different types of convolution layers including standard convolution, dilated convolution, transposed convolution, point-wise convolution, and DSC. With flexible architecture, it achieves 17.97 × higher throughput with DTN and DSC layer fusion architecture reduces 34.7% of overall external memory access. Implemented with 28nm CMOS technology, the unified CNN processor shows 43.6 mW power consumption and 4.94 TOPS/W energy efficiency. As a result, the proposed system with DTN realizes 40.07 frames-per-second (fps) throughputs in multi-object semantic segmentation application with high resolution driving scenes dataset. | - |
dc.identifier.bibliographicCitation | IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS | - |
dc.identifier.doi | 10.1109/TCSI.2024.3349588 | - |
dc.identifier.issn | 1549-8328 | - |
dc.identifier.scopusid | 2-s2.0-85182951628 | - |
dc.identifier.uri | https://scholarworks.unist.ac.kr/handle/201301/81648 | - |
dc.identifier.wosid | 001167331800001 | - |
dc.language | 영어 | - |
dc.publisher | Institute of Electrical and Electronics Engineers | - |
dc.title | An Energy-Efficient, Unified CNN Accelerator for Real-Time Multi-Object Semantic Segmentation for Autonomous Vehicle | - |
dc.type | Article | - |
dc.description.isOpenAccess | FALSE | - |
dc.relation.journalWebOfScienceCategory | Engineering, Electrical & Electronic | - |
dc.relation.journalResearchArea | Engineering | - |
dc.type.docType | Article | - |
dc.description.journalRegisteredClass | scie | - |
dc.description.journalRegisteredClass | scopus | - |
dc.subject.keywordAuthor | autonomous electric vehicle system | - |
dc.subject.keywordAuthor | Convolution | - |
dc.subject.keywordAuthor | Convolutional neural network | - |
dc.subject.keywordAuthor | Convolutional neural networks | - |
dc.subject.keywordAuthor | depth-wise separable convolution | - |
dc.subject.keywordAuthor | dilated convolution | - |
dc.subject.keywordAuthor | Energy efficiency | - |
dc.subject.keywordAuthor | Feature extraction | - |
dc.subject.keywordAuthor | Kernel | - |
dc.subject.keywordAuthor | multi-object semantic segmentation | - |
dc.subject.keywordAuthor | Real-time systems | - |
dc.subject.keywordAuthor | Semantic segmentation | - |
dc.subject.keywordAuthor | transposed convolution | - |
dc.subject.keywordAuthor | trilateral network | - |
dc.subject.keywordPlus | NEURAL-NETWORK | - |
Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.
Tel : 052-217-1404 / Email : scholarworks@unist.ac.kr
Copyright (c) 2023 by UNIST LIBRARY. All rights reserved.
ScholarWorks@UNIST was established as an OAK Project for the National Library of Korea.