14.6 A 0.62mW ultra-low-power convolutional-neural-network face-recognition processor and a CIS integrated with always-on haar-like face detector. Bong, K., Choi, S., Kim, C., Kang, S., Kim, Y., & Yoo, H. J. In 2017 IEEE International Solid-State Circuits Conference (ISSCC), pages 248–249, February, 2017. doi abstract bibtex Recently, face recognition (FR) based on always-on CIS has been investigated for the next-generation UI/UX of wearable devices. A FR system, shown in Fig. 14.6.1, was developed as a life-cycle analyzer or a personal black box, constantly recording the people we meet, along with time and place information. In addition, FR with always-on capability can be used for user authentication for secure access to his or her smart phone and other personal systems. Since wearable devices have a limited battery capacity for a small form factor, extremely low power consumption is required, while maintaining high recognition accuracy. Previously, a 23mW FR accelerator [1] was proposed, but its accuracy was low due to its hand-crafted feature-based algorithm. Deep learning using a convolutional neural network (CNN) is essential to achieve high accuracy and to enhance device intelligence. However, previous CNN processors (CNNP) [2-3] consume too much power, resulting in \textless;10 hours operation time with a 190mAh coin battery.
@inproceedings{bong_14.6_2017,
title = {14.6 {A} 0.62mW ultra-low-power convolutional-neural-network face-recognition processor and a {CIS} integrated with always-on haar-like face detector},
doi = {10.1109/ISSCC.2017.7870354},
abstract = {Recently, face recognition (FR) based on always-on CIS has been investigated for the next-generation UI/UX of wearable devices. A FR system, shown in Fig. 14.6.1, was developed as a life-cycle analyzer or a personal black box, constantly recording the people we meet, along with time and place information. In addition, FR with always-on capability can be used for user authentication for secure access to his or her smart phone and other personal systems. Since wearable devices have a limited battery capacity for a small form factor, extremely low power consumption is required, while maintaining high recognition accuracy. Previously, a 23mW FR accelerator [1] was proposed, but its accuracy was low due to its hand-crafted feature-based algorithm. Deep learning using a convolutional neural network (CNN) is essential to achieve high accuracy and to enhance device intelligence. However, previous CNN processors (CNNP) [2-3] consume too much power, resulting in {\textless};10 hours operation time with a 190mAh coin battery.},
booktitle = {2017 {IEEE} {International} {Solid}-{State} {Circuits} {Conference} ({ISSCC})},
author = {Bong, K. and Choi, S. and Kim, C. and Kang, S. and Kim, Y. and Yoo, H. J.},
month = feb,
year = {2017},
keywords = {Important},
pages = {248--249}
}
Downloads: 0
{"_id":"aDmGnafqdrnr3f3e2","bibbaseid":"bong-choi-kim-kang-kim-yoo-146a062mwultralowpowerconvolutionalneuralnetworkfacerecognitionprocessorandacisintegratedwithalwaysonhaarlikefacedetector-2017","authorIDs":[],"author_short":["Bong, K.","Choi, S.","Kim, C.","Kang, S.","Kim, Y.","Yoo, H. J."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","title":"14.6 A 0.62mW ultra-low-power convolutional-neural-network face-recognition processor and a CIS integrated with always-on haar-like face detector","doi":"10.1109/ISSCC.2017.7870354","abstract":"Recently, face recognition (FR) based on always-on CIS has been investigated for the next-generation UI/UX of wearable devices. A FR system, shown in Fig. 14.6.1, was developed as a life-cycle analyzer or a personal black box, constantly recording the people we meet, along with time and place information. In addition, FR with always-on capability can be used for user authentication for secure access to his or her smart phone and other personal systems. Since wearable devices have a limited battery capacity for a small form factor, extremely low power consumption is required, while maintaining high recognition accuracy. Previously, a 23mW FR accelerator [1] was proposed, but its accuracy was low due to its hand-crafted feature-based algorithm. Deep learning using a convolutional neural network (CNN) is essential to achieve high accuracy and to enhance device intelligence. However, previous CNN processors (CNNP) [2-3] consume too much power, resulting in \\textless;10 hours operation time with a 190mAh coin battery.","booktitle":"2017 IEEE International Solid-State Circuits Conference (ISSCC)","author":[{"propositions":[],"lastnames":["Bong"],"firstnames":["K."],"suffixes":[]},{"propositions":[],"lastnames":["Choi"],"firstnames":["S."],"suffixes":[]},{"propositions":[],"lastnames":["Kim"],"firstnames":["C."],"suffixes":[]},{"propositions":[],"lastnames":["Kang"],"firstnames":["S."],"suffixes":[]},{"propositions":[],"lastnames":["Kim"],"firstnames":["Y."],"suffixes":[]},{"propositions":[],"lastnames":["Yoo"],"firstnames":["H.","J."],"suffixes":[]}],"month":"February","year":"2017","keywords":"Important","pages":"248–249","bibtex":"@inproceedings{bong_14.6_2017,\n\ttitle = {14.6 {A} 0.62mW ultra-low-power convolutional-neural-network face-recognition processor and a {CIS} integrated with always-on haar-like face detector},\n\tdoi = {10.1109/ISSCC.2017.7870354},\n\tabstract = {Recently, face recognition (FR) based on always-on CIS has been investigated for the next-generation UI/UX of wearable devices. A FR system, shown in Fig. 14.6.1, was developed as a life-cycle analyzer or a personal black box, constantly recording the people we meet, along with time and place information. In addition, FR with always-on capability can be used for user authentication for secure access to his or her smart phone and other personal systems. Since wearable devices have a limited battery capacity for a small form factor, extremely low power consumption is required, while maintaining high recognition accuracy. Previously, a 23mW FR accelerator [1] was proposed, but its accuracy was low due to its hand-crafted feature-based algorithm. Deep learning using a convolutional neural network (CNN) is essential to achieve high accuracy and to enhance device intelligence. However, previous CNN processors (CNNP) [2-3] consume too much power, resulting in {\\textless};10 hours operation time with a 190mAh coin battery.},\n\tbooktitle = {2017 {IEEE} {International} {Solid}-{State} {Circuits} {Conference} ({ISSCC})},\n\tauthor = {Bong, K. and Choi, S. and Kim, C. and Kang, S. and Kim, Y. and Yoo, H. J.},\n\tmonth = feb,\n\tyear = {2017},\n\tkeywords = {Important},\n\tpages = {248--249}\n}\n\n","author_short":["Bong, K.","Choi, S.","Kim, C.","Kang, S.","Kim, Y.","Yoo, H. J."],"key":"bong_14.6_2017","id":"bong_14.6_2017","bibbaseid":"bong-choi-kim-kang-kim-yoo-146a062mwultralowpowerconvolutionalneuralnetworkfacerecognitionprocessorandacisintegratedwithalwaysonhaarlikefacedetector-2017","role":"author","urls":{},"keyword":["Important"],"downloads":0},"bibtype":"inproceedings","biburl":"https://bibbase.org/zotero/ky25","creationDate":"2019-05-11T17:47:04.390Z","downloads":0,"keywords":["important"],"search_terms":["62mw","ultra","low","power","convolutional","neural","network","face","recognition","processor","cis","integrated","always","haar","face","detector","bong","choi","kim","kang","kim","yoo"],"title":"14.6 A 0.62mW ultra-low-power convolutional-neural-network face-recognition processor and a CIS integrated with always-on haar-like face detector","year":2017,"dataSources":["XxiQtwZYfozhQmvGR"]}