DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Yoo, Hoi-Jun | - |
dc.contributor.advisor | 유회준 | - |
dc.contributor.author | Kim, Sangyeob | - |
dc.date.accessioned | 2021-05-12T19:32:58Z | - |
dc.date.available | 2021-05-12T19:32:58Z | - |
dc.date.issued | 2020 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=901534&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/283802 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 전기및전자공학부, 2020.2,[1책 :] | - |
dc.description.abstract | A low power face recognition (FR) convolutional neural network (CNN) processor is proposed with high power efficiency to achieve always-on FR in mobile devices. Three key features enable a power-efficient FR CNN. First, tile-based clustering (THC) is proposed for reducing the computation overhead of hierarchical clustering Second, a low latency tile-based hierarchical clustering core is proposed. It supports an approximated clustering method that removes distance updates and increases pipeline utilization by up to 98.6%. Finally, a similar feature skipping binary convolution core with separated accumulation cores is proposed to increase power efficiency through skipping redundant MAC operation between duplicated input features and weights. It can reduce 35.8% of the total computation considering both CNN and THC. The proposed features enable the proposed processor to operate FR CNN with 17.3 TOPS/W power efficiency, which is 1.3× higher compared to the previous state-of-the-art FR CNN processor. Implemented in 65nm CMOS technology, the 6mm2 FR CNN processor shows 0.5mW power consumption at 1 frames-per-second always-on face recognition. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Face recognition▼aConvolutional neural network▼aTile-based processing▼aHierarchical clustering▼aActivation skipping | - |
dc.subject | 얼굴 인식▼a심층신경망▼a타일 기반 연산▼a클러스터링▼a입력 기반 연산 생략 | - |
dc.title | (An) energy-efficient CNN accelerator with similar feature skipping for face recognition in mobile devices | - |
dc.title.alternative | 모바일 얼굴 인식을 위한 에너지 효율적인 유사 특징 연산 생략 심층 신경망 가속기 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :전기및전자공학부, | - |
dc.contributor.alternativeauthor | 김상엽 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.