A 46M 512512 Ultralow Power Stacked Digital Pixel Sensor With Triple Quantization And 127Db Dynamic Range

• doi: 10.1109/IEDM13553.2020.9371913
• title: A 4.6μm, 512×512, Ultra-Low Power Stacked Digital Pixel Sensor with Triple Quantization and 127dB Dynamic Range
• publisher: IEEE
• isbn: 978-1-7281-8889-8
• issn: 0163-1918
• rank: 445
• access_type: LOCKED
• content_type: Conferences
• abstract: We present a global shutter (GS), digital pixel sensor (DPS) that leverages stacked CMOS image sensor (CIS) technology to meet the ultra-low power, ultra-wide dynamic range (DR) requirements for battery-powered, always-on mobile computer vision (CV) applications. The DPS pixel is partitioned between two silicon layers via pixel-level connections, has an in-pixel ADC, 10-bit SRAM, and 4.6µm pitch. We introduce a triple quantization (3Q) scheme that combines a time-to-saturation (TTS) quantization mode and two linear ADC modes within a single exposure to achieve 127dB intra-scene DR. The sensor has a 512×512 effective resolution and consumes 5.3mW in 3Q operation at 30fps.
• article_number: 9371913
• pdf_url: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9371913
• html_url: https://ieeexplore.ieee.org/document/9371913/
• abstract_url: https://ieeexplore.ieee.org/document/9371913/
• publication_title: 2020 IEEE International Electron Devices Meeting (IEDM)
• conference_location: San Francisco, CA, USA
• conference_dates: 12-18 Dec. 2020
• publication_number: 9371868
• is_number: 9371888
• publication_year: 2020
• publication_date: 12-18 Dec. 2020
• start_page: 16.1.1
• end_page: 16.1.4
• citing_paper_count: 7
• citing_patent_count: 0
• download_count: 2291
• insert_date: 20210311
• index_terms:
  • ieee_terms:
    • Performance evaluation
    • Quantization (signal)
    • Random access memory
    • Dynamic range
    • Silicon
    • Timing
    • Low-power electronics
  • dynamic_index_terms:
    • Ultra-low Power
    • Sensor Pixel
    • Image Sensor
    • Camera Sensor
    • Single Exposure
    • Computer Vision Applications
    • Quantization Scheme
    • Quantification Strategy
    • Μm Pitch
    • Mobile Devices
    • Portable Devices
    • Power Consumption
    • Bottom Layer
    • Low Light
    • Near-infrared
    • Light Levels
    • Lower Gain
    • Shot Noise
    • Poisson Noise
    • Shot-noise
    • High Power Consumption
    • Electron Counting
    • Digital Counting
    • Quantization Noise
    • Conversion Gain
    • Linear Gain
    • Test Chart
• isbn_formats:
  • format: Print on Demand(PoD) ISBN, value: 978-1-7281-8889-8, isbnType: New-2005
  • format: Electronic ISBN, value: 978-1-7281-8888-1, isbnType: New-2005
• authors:
  • Author Name: Chiao Liu
    Affiliation: Facebook Reality Labs, Facebook Inc, Menlo Park, CA, USA
    Author URL: https://ieeexplore.ieee.org/author/37087882293
    ID: 37087882293
    Order: 1
    Author Affiliations:
    • Facebook Reality Labs, Facebook Inc, Menlo Park, CA, USA
  • Author Name: Lyle Bainbridge
    Affiliation: Facebook Reality Labs, Facebook Inc, Menlo Park, CA, USA
    Author URL: https://ieeexplore.ieee.org/author/37088866945
    ID: 37088866945
    Order: 2
    Author Affiliations:
    • Facebook Reality Labs, Facebook Inc, Menlo Park, CA, USA
  • Author Name: Andrew Berkovich
    Affiliation: Facebook Reality Labs, Facebook Inc, Menlo Park, CA, USA
    Author URL: https://ieeexplore.ieee.org/author/38234682700
    ID: 38234682700
    Order: 3
    Author Affiliations:
    • Facebook Reality Labs, Facebook Inc, Menlo Park, CA, USA
  • Author Name: Song Chen
    Affiliation: Facebook Reality Labs, Facebook Inc, Menlo Park, CA, USA
    Author URL: https://ieeexplore.ieee.org/author/37089096400
    ID: 37089096400
    Order: 4
    Author Affiliations:
    • Facebook Reality Labs, Facebook Inc, Menlo Park, CA, USA
  • Author Name: Wei Gao
    Affiliation: Facebook Reality Labs, Facebook Inc, Menlo Park, CA, USA
    Author URL: https://ieeexplore.ieee.org/author/37088866847
    ID: 37088866847
    Order: 5
    Author Affiliations:
    • Facebook Reality Labs, Facebook Inc, Menlo Park, CA, USA
  • Author Name: Tsung-Hsun Tsai
    Affiliation: Facebook Reality Labs, Facebook Inc, Menlo Park, CA, USA
    Author URL: https://ieeexplore.ieee.org/author/38102063500
    ID: 38102063500
    Order: 6
    Author Affiliations:
    • Facebook Reality Labs, Facebook Inc, Menlo Park, CA, USA
  • Author Name: Kazuya Mori
    Affiliation: Brillnics Japan Inc., Tokyo, Japan
    Author URL: https://ieeexplore.ieee.org/author/37088868065
    ID: 37088868065
    Order: 7
    Author Affiliations:
    • Brillnics Japan Inc., Tokyo, Japan
  • Author Name: Rimon Ikeno
    Affiliation: Brillnics Japan Inc., Tokyo, Japan
    Author URL: https://ieeexplore.ieee.org/author/38027564500
    ID: 38027564500
    Order: 8
    Author Affiliations:
    • Brillnics Japan Inc., Tokyo, Japan
  • Author Name: Masayuki Uno
    Affiliation: Brillnics Japan Inc., Tokyo, Japan
    Author URL: https://ieeexplore.ieee.org/author/37088866327
    ID: 37088866327
    Order: 9
    Author Affiliations:
    • Brillnics Japan Inc., Tokyo, Japan
  • Author Name: Toshiyuki Isozaki
    Affiliation: Brillnics Japan Inc., Tokyo, Japan
    Author URL: https://ieeexplore.ieee.org/author/37088867154
    ID: 37088867154
    Order: 10
    Author Affiliations:
    • Brillnics Japan Inc., Tokyo, Japan
  • Author Name: Yu-Lin Tsai
    Affiliation: Brillnics Inc., Hsinchu, Taiwan
    Author URL: https://ieeexplore.ieee.org/author/37088867839
    ID: 37088867839
    Order: 11
    Author Affiliations:
    • Brillnics Inc., Hsinchu, Taiwan
  • Author Name: Isao Takayanagi
    Affiliation: Brillnics Japan Inc., Tokyo, Japan
    Author URL: https://ieeexplore.ieee.org/author/37564618500
    ID: 37564618500
    Order: 12
    Author Affiliations:
    • Brillnics Japan Inc., Tokyo, Japan
  • Author Name: Junichi Nakamura
    Affiliation: Brillnics Japan Inc., Tokyo, Japan
    Author URL: https://ieeexplore.ieee.org/author/37564387800
    ID: 37564387800
    Order: 13
    Author Affiliations:
    • Brillnics Japan Inc., Tokyo, Japan

Image Sensor

• sensor_type: CMOS
• resolution: 512x512
• dynamic_range: 127dB
• pixel_size: 4.6μm
• 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: 30fps
• power_consumption: 5.3mW
• noise: 4.2e-

Applications & Benefits

• cell_imaging: Not specified
• benefits: Ultra-low power, ultra-wide dynamic range suitable for battery-powered, always-on mobile computer vision applications

Supporting Organizations

• supported_by: Facebook Reality Labs, Brillnics Japan Inc., Brillnics Inc.

Manuscript Details

• publication_date: 12-18 Dec. 2020

Relevancy Score

• score: 10

Processed JSON Filename

request_0da92285-6f79-48fa-9930-7957788b1eb7-a_46m_512512_ultralow_power_stacked_digital_pixel_sensor_with_triple_quantization_and_127db_dynamic_range.json