Ai Socs For Arvr Userinteraction

• doi: 10.1109/IEDM19574.2021.9720630
• title: AI SoCs for AR/VR User-Interaction
• publisher: IEEE
• isbn: 978-1-6654-2573-5
• issn: 0163-1918
• rank: 1355
• access_type: LOCKED
• content_type: Conferences
• abstract: AR/VR systems require heavy processing for object recognition, voice recognition, hand gesture recognition (HGR), and camera pose estimation with limited battery and computation capability. CMOS Image Sensor (CIS) on AR/VR systems can integrate functions to reduce data transaction and power consumption. 3 Functional CIS and a UI processor are explained; ‘Eye-Mouse’ with gaze tracking, Event-driven ultra-low-power face detection, video-based human action recognition, and CNN-based HGR. AR/VR application processors are also explained with their architecture and featured functions. In the future of AR/VR user interaction, SoC will utilize more DNN functional blocks for 3D Point Neural Network and fusion of more sensors for better accuracy, lower power consumption, and easy implementation.
• article_number: 9720630
• pdf_url: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9720630
• html_url: https://ieeexplore.ieee.org/document/9720630/
• abstract_url: https://ieeexplore.ieee.org/document/9720630/
• publication_title: 2021 IEEE International Electron Devices Meeting (IEDM)
• conference_location: San Francisco, CA, USA
• conference_dates: 11-16 Dec. 2021
• publication_number: 9720433
• is_number: 9720494
• publication_year: 2021
• publication_date: 11-16 Dec. 2021
• start_page: 35.4.1
• end_page: 35.4.4
• citing_paper_count: 3
• citing_patent_count: 0
• download_count: 800
• insert_date: 20220309
• index_terms:
  • ieee_terms:
    • Power demand
    • Three-dimensional displays
    • Image recognition
    • Pose estimation
    • Neural networks
    • Speech recognition
    • Gesture recognition
  • dynamic_index_terms:
    • Human-computer Interaction
    • Human Computer Interaction
    • User Interaction
    • Low Power
    • Power Consumption
    • Eye-tracking
    • Object Recognition
    • Action Recognition
    • Pose Estimation
    • Face Detection
    • Gesture Recognition
    • Camera Pose
    • Voice Recognition
    • Human Activity Recognition
    • Camera Pose Estimation
    • Hand Gesture Recognition
    • User Experience
    • Edge Detection
    • Accurate 3D
    • Virtual Objects
    • RGB Data
    • Stereo Matching
    • Hand Tracking
    • Finger Tracking
    • Detection Circuit
    • Motion Events
    • 65-nm CMOS
    • Frame Resolution
    • Resolution Of Each Frame
• isbn_formats:
  • format: Print on Demand(PoD) ISBN, value: 978-1-6654-2573-5, isbnType: New-2005
  • format: Electronic ISBN, value: 978-1-6654-2572-8, isbnType: New-2005
• authors:
  • Author Name: Junha Ryu
    Affiliation: Korea Advanced Institute of Science and Technology, Daejeon, Korea
    Author URL: https://ieeexplore.ieee.org/author/37088852779
    ID: 37088852779
    Order: 1
    Author Affiliations:
    • Korea Advanced Institute of Science and Technology, Daejeon, Korea
  • Author Name: Dongseok Im
    Affiliation: Korea Advanced Institute of Science and Technology, Daejeon, Korea
    Author URL: https://ieeexplore.ieee.org/author/37086885443
    ID: 37086885443
    Order: 2
    Author Affiliations:
    • Korea Advanced Institute of Science and Technology, Daejeon, Korea
  • Author Name: Hoi-Jun Yoo
    Affiliation: Korea Advanced Institute of Science and Technology, Daejeon, Korea
    Author URL: https://ieeexplore.ieee.org/author/37274307200
    ID: 37274307200
    Order: 3
    Author Affiliations:
    • Korea Advanced Institute of Science and Technology, Daejeon, Korea

Image Sensor

• sensor_type: CMOS
• resolution:
• dynamic_range: Not specified
• pixel_size: Not specified
• dark_current: Not specified

Optical Data

• focal_length: Not specified
• aperture: Not specified
• field_of_view: Not specified
• distortion: Not specified

Performance Metrics

• frame_rate: <10 fps as per traditional AP
• power_consumption: >1 W for traditional AP, 24 µW for ultra-low power FD SoC, 0.82 µW for HAR SoC, <10 mW for 3D HGR processor, 27 mW for CPE architecture

Applications & Benefits

• cell_imaging: Not specified
• benefits: Better accuracy, lower power consumption, and easy implementation in AR/VR devices.

Supporting Organizations

• supported_by: Korea Advanced Institute of Science and Technology

Manuscript Details

• publication_date: 11-16 Dec. 2021

Relevancy Score

• score: 9
• missing_fields:
  • resolution
  • dynamic_range
  • pixel_size
  • dark_current
  • focal_length
  • aperture
  • field_of_view
  • distortion
  • signal_to_noise_ratio
  • sensitivity
  • shutter_speed
  • noise

Processed JSON Filename

request_565ff44b-7275-40f4-9928-ec5e3c7e776f-ai_socs_for_arvr_userinteraction.json