A 271 Njpixel Gazeactivated Object Recognition System For Lowpower Mobile Smart Glasses

• doi: 10.1109/JSSC.2015.2476786
• title: A 2.71 nJ/Pixel Gaze-Activated Object Recognition System for Low-Power Mobile Smart Glasses
• publisher: IEEE
• isbn:
• issn: 1558-173X
• rank: 349
• access_type: LOCKED
• content_type: Journals
• abstract: A low-power object recognition (OR) system with intuitive gaze user interface (UI) is proposed for battery-powered smart glasses. For low-power gaze UI, we propose a low-power single-chip gaze estimation sensor, called gaze image sensor (GIS). In GIS, a novel column-parallel pupil edge detection circuit (PEDC) with new pupil edge detection algorithm XY pupil detection (XY-PD) is proposed which results in 2.9× power reduction with 16× larger resolution compared to previous work. Also, a logarithmic SIMD processor is proposed for robust pupil center estimation, (1 pixel error, with low-power floating-point implementation. For OR, low-power multicore OR processor (ORP) is implemented. In ORP, task-level pipeline with keypoint-level scoring is proposed to reduce the number of cores as well as the operating frequency of keypoint-matching processor (KMP) for low-power consumption. Also, dual-mode convolutional neural network processor (CNNP) is designed for fast tile selection without external memory accesses. In addition, a pipelined descriptor generation processor (DGP) with LUT-based nonlinear operation is newly proposed for low-power OR. Lastly, dynamic voltage and frequency scaling (DVFS) for dynamic power reduction in ORP is applied. Combining both of the GIS and ORP fabricated in 65 nm CMOS logic process, only 75 mW average power consumption is achieved with real-time OR performance, which is 1.2× and 4.4× lower power than the previously published work.
• article_number: 7286768
• pdf_url: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7286768
• html_url: https://ieeexplore.ieee.org/document/7286768/
• abstract_url: https://ieeexplore.ieee.org/document/7286768/
• publication_title: IEEE Journal of Solid-State Circuits
• conference_location:
• conference_dates:
• publication_number: 4
• is_number: 7368951
• publication_year: 2016
• publication_date: Jan. 2016
• start_page: 45
• end_page: 55
• citing_paper_count: 13
• citing_patent_count: 0
• download_count: 2309
• insert_date: 20151001
• index_terms:
  • ieee_terms:
    • Image edge detection
    • Glass
    • Estimation
    • Robustness
    • Acceleration
    • Pipelines
    • Calibration
  • author_terms:
    • Convolutional neural network (CNN)
    • dynamic voltage and frequency scaling (DVFS)
    • eye tracking
    • focal-plane processing
    • gaze estimation
    • logarithmic approximation
    • object recognition (OR)
    • smart glasses
    • vision chip
    • Convolutional neural network (CNN)
    • dynamic voltage and frequency scaling (DVFS)
    • eye tracking
    • focal-plane processing
    • gaze estimation
    • logarithmic approximation
    • object recognition (OR)
    • smart glasses
    • vision chip
  • dynamic_index_terms:
    • Object Recognition
    • Object Recognition System
    • Low Power
    • Convolutional Neural Network
    • Power Consumption
    • Multi-core
    • Quad-core
    • Dual-core
    • CPU Cores
    • Operating Frequency
    • Image Sensor
    • Camera Sensor
    • Edge Detection
    • Reduction In Power
    • Frequency Scale
    • Dynamic Power
    • Dynamic Voltage
    • External Memory
    • Edge Detection Algorithm
    • Detection Circuit
    • Pupil Center
    • Nonlinear Function
    • Nonlinear Mapping
    • Eye-tracking
    • Multilayer Perceptron
    • Ellipse Fitting
    • Multilayer Perceptron Classifier
    • Division Operation
    • Image Tiles
    • Gaze Points
    • Target Object
    • Eyelashes
    • Calibration Step
    • Head-mounted Display
    • Query Image
• authors:
  • Author Name: Injoon Hong
    Affiliation: Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
    Author URL: https://ieeexplore.ieee.org/author/38242546400
    ID: 38242546400
    Order: 1
    Author Affiliations:
    • Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
  • Author Name: Kyeongryeol Bong
    Affiliation: Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
    Author URL: https://ieeexplore.ieee.org/author/37085520085
    ID: 37085520085
    Order: 2
    Author Affiliations:
    • Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
  • Author Name: Dongjoo Shin
    Affiliation: Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
    Author URL: https://ieeexplore.ieee.org/author/37085435042
    ID: 37085435042
    Order: 3
    Author Affiliations:
    • Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
  • Author Name: Seongwook Park
    Affiliation: Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
    Author URL: https://ieeexplore.ieee.org/author/37067908100
    ID: 37067908100
    Order: 4
    Author Affiliations:
    • Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
  • Author Name: Kyuho Jason Lee
    Affiliation: Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
    Author URL: https://ieeexplore.ieee.org/author/37085725206
    ID: 37085725206
    Order: 5
    Author Affiliations:
    • Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
  • Author Name: Youchang Kim
    Affiliation: Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
    Author URL: https://ieeexplore.ieee.org/author/38580636400
    ID: 38580636400
    Order: 6
    Author Affiliations:
    • Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
  • Author Name: Hoi-Jun Yoo
    Affiliation: Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
    Author URL: https://ieeexplore.ieee.org/author/37274307200
    ID: 37274307200
    Order: 7
    Author Affiliations:
    • Department of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea

Image Sensor

• sensor_type: CMOS
• resolution: 320x240
• 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: 30 fps
• power_consumption: 75 mW

Applications & Benefits

• cell_imaging: Used for eye tracking and object recognition in smart glasses.
• benefits: Low power consumption, robust pupil center detection, supports gaze-activated user interface.

Supporting Organizations

• supported_by: Korea Advanced Institute of Science and Technology (KAIST)

Manuscript Details

• publication_date: Jan. 2016

Relevancy Score

• score: 10
• missing_fields:
  • fill factor
  • quantum efficiency
  • readout speed
  • noise sources
  • analog-to-digital conversion techniques
  • microlenses
  • on-chip colour filters
  • global or rolling shutters
  • backside illumination

Processed JSON Filename

request_356f524b-beaf-4163-8ad5-25105141c30e-a_271_njpixel_gazeactivated_object_recognition_system_for_lowpower_mobile_smart_glasses.json