DC Field | Value | Language |
---|---|---|
dc.contributor.author | Bong, Kyeongryeol | ko |
dc.contributor.author | Choi, Sungpill | ko |
dc.contributor.author | Kim, Changhyeon | ko |
dc.contributor.author | Han, Donghyeon | ko |
dc.contributor.author | Yoo, Hoi-Jun | ko |
dc.date.accessioned | 2018-01-30T05:48:54Z | - |
dc.date.available | 2018-01-30T05:48:54Z | - |
dc.date.created | 2018-01-15 | - |
dc.date.created | 2018-01-15 | - |
dc.date.issued | 2018-01 | - |
dc.identifier.citation | IEEE JOURNAL OF SOLID-STATE CIRCUITS, v.53, no.1, pp.115 - 123 | - |
dc.identifier.issn | 0018-9200 | - |
dc.identifier.uri | http://hdl.handle.net/10203/239456 | - |
dc.description.abstract | A Low-power convolutional neural network (CNN)-based face recognition system is proposed for the user authentication in smart devices. The system consists of two chips: an always-on CMOS image sensor (CIS)-based face detector (FD) and a low-power CNN processor. For always-on FD, analog-digital Hybrid Haar-like FD is proposed to improve the energy efficiency of FD by 39%. For low-power CNN processing, the CNN processor with 1024 MAC units and 8192-bit-wide local distributed memory operates at near threshold voltage, 0.46 V with 5-MHz clock frequency. In addition, the separable filter approximation is adopted for the workload reduction of CNN, and transpose-read SRAM using 7T SRAM cell is proposed to reduce the activity factor of the data read operation. Implemented in 65-nm CMOS technology, the 3.30 x 3.36 mm(2) CIS chip and the 4 x 4 mm(2) CNN processor consume 0.62 mW to evaluate one face at 1 fps and achieved 97% accuracy in LFW dataset. | - |
dc.language | English | - |
dc.publisher | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC | - |
dc.title | A Low-Power Convolutional Neural Network Face Recognition Processor and a CIS Integrated With Always-on Face Detector | - |
dc.type | Article | - |
dc.identifier.wosid | 000418873800010 | - |
dc.identifier.scopusid | 2-s2.0-85038850855 | - |
dc.type.rims | ART | - |
dc.citation.volume | 53 | - |
dc.citation.issue | 1 | - |
dc.citation.beginningpage | 115 | - |
dc.citation.endingpage | 123 | - |
dc.citation.publicationname | IEEE JOURNAL OF SOLID-STATE CIRCUITS | - |
dc.identifier.doi | 10.1109/JSSC.2017.2767705 | - |
dc.contributor.localauthor | Yoo, Hoi-Jun | - |
dc.contributor.nonIdAuthor | Han, Donghyeon | - |
dc.description.isOpenAccess | N | - |
dc.type.journalArticle | Article; Proceedings Paper | - |
dc.subject.keywordAuthor | Always-on system | - |
dc.subject.keywordAuthor | convolutional neural network (CNN) | - |
dc.subject.keywordAuthor | face recognition (FR) | - |
dc.subject.keywordAuthor | functional CMOS image sensor (CIS) | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.