There are no files associated with this item.
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.citation.conferencePlace | US | - |
dc.citation.conferencePlace | San Francisco | - |
dc.citation.endPage | 222 | - |
dc.citation.startPage | 220 | - |
dc.citation.title | IEEE International Solid-State Circuits Conference | - |
dc.contributor.author | Choi, Sungpill | - |
dc.contributor.author | Lee, Jinsu | - |
dc.contributor.author | Lee, Kyuho | - |
dc.contributor.author | Yoo, Hoi-Jun | - |
dc.date.accessioned | 2023-12-19T17:37:15Z | - |
dc.date.available | 2023-12-19T17:37:15Z | - |
dc.date.created | 2018-08-07 | - |
dc.date.issued | 2018-02-11 | - |
dc.description.abstract | Recently, 3D hand-gesture recognition (HGR) has become an important feature in smart mobile devices, such as head-mounted displays (HMDs) or smartphones for AR/VR applications. A 3D HGR system in Fig. 13.4.1 enables users to interact with virtual 3D objects using depth sensing and hand tracking. However, a previous 3D HGR system, such as Hololens [1], utilized a power consuming time-of-flight (ToF) depth sensor (>2W) limiting 3D HGR operation to less than 3 hours. Even though stereo matching was used instead of ToF for depth sensing with low power consumption [2], it could not provide interaction with virtual 3D objects because depth information was used only for hand segmentation. The HGR-based UI system in smart mobile devices, such as HMDs, must be low power consumption (<;10mW), while maintaining real-time operation (<;33.3ms). A convolutional neural network (CNN) can be adopted to enhance the accuracy of the low-power stereo matching. The CNN-based HGR system comprises two 6-layer CNNs (stereo) without any pooling layers to preserve geometrical information and an iterative-closest-point/particle-swarm optimization-based (ICP-PSO) hand tracking to acquire 3D coordinates of a user's fingertips and palm from the hand depth. The CNN learns the skin color and texture to detect the hand accurately, comparable to ToF, in the low-power stereo matching system irrespective of variations in external conditions [3]. However, it requires >1000 more MAC operations than previous feature-based stereo depth sensing, which is difficult in real-time with a mobile CPU, and therefore, a dedicated low-power CNN-based stereo matching SoC is required. | - |
dc.identifier.bibliographicCitation | IEEE International Solid-State Circuits Conference, pp.220 - 222 | - |
dc.identifier.doi | 10.1109/ISSCC.2018.8310263 | - |
dc.identifier.issn | 0193-6530 | - |
dc.identifier.scopusid | 2-s2.0-85046497098 | - |
dc.identifier.uri | https://scholarworks.unist.ac.kr/handle/201301/35080 | - |
dc.identifier.url | https://ieeexplore.ieee.org/document/8310263/ | - |
dc.language | 영어 | - |
dc.publisher | 65th IEEE International Solid-State Circuits Conference, ISSCC 2018 | - |
dc.title | A 9.02mW CNN-stereo-based real-time 3D hand-gesture recognition processor for smart mobile devices | - |
dc.type | Conference Paper | - |
dc.date.conferenceDate | 2018-02-11 | - |
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.