Learning Superresolved Depth From Active Gated Imaging

• doi: 10.1109/ITSC.2018.8569590
• title: Learning Super-resolved Depth from Active Gated Imaging
• publisher: IEEE
• isbn: 978-1-7281-0324-2
• issn: 2153-0009
• rank: 2868
• access_type: LOCKED
• content_type: Conferences
• abstract: Environment perception for autonomous driving is doomed by the trade-off between range-accuracy and resolution: current sensors that deliver very precise depth information are usually restricted to low resolution because of technology or cost limitations. In this work, we exploit depth information from an active gated imaging system based on cost-sensitive diode and CMOS technology. Learning a mapping between pixel intensities of three gated slices and depth produces a super-resolved depth map image with respectable relative accuracy of 5 % in between 25-80 m. By design, depth information is perfectly aligned with pixel intensity values.
• article_number: 8569590
• pdf_url: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8569590
• html_url: https://ieeexplore.ieee.org/document/8569590/
• abstract_url: https://ieeexplore.ieee.org/document/8569590/
• publication_title: 2018 21st International Conference on Intelligent Transportation Systems (ITSC)
• conference_location: Maui, HI, USA
• conference_dates: 4-7 Nov. 2018
• publication_number: 8543039
• is_number: 8569013
• publication_year: 2018
• publication_date: 4-7 Nov. 2018
• start_page: 3051
• end_page: 3058
• citing_paper_count: 9
• citing_patent_count: 0
• download_count: 254
• insert_date: 20181209
• index_terms:
  • ieee_terms:
    • Logic gates
    • Delays
    • Imaging
    • Shape
    • Lasers
    • Economic indicators
    • Correlation
  • dynamic_index_terms:
    • Pixel Intensity
    • Depth Map
    • Function Of Depth
    • Depth Information
    • Current Sensor
    • Complementary Metal Oxide Semiconductor Technology
    • Complementary Metal-oxide-semiconductor Technology
    • Neural Network
    • Illumination
    • Lighting
    • Training Set
    • Convolutional Neural Network
    • Nonlinear Function
    • Nonlinear Mapping
    • Laser Pulse
    • Grid Search
    • Image Sensor
    • Camera Sensor
    • Depth Estimation
    • Pulse Shape
    • Bad Weather
    • Shadowing Effect
    • Single Hidden Layer
    • Single-hidden Layer
    • Night Vision
    • Night-vision
    • Lidar System
    • Baseline Algorithms
    • Simple Neural Network
    • Gain Modulation
    • Mean Absolute Relative Error
    • Saturated Pixels
    • Maximum And Minimum
    • Stationary Point
    • Extreme Values
    • Extremal
    • Training Data
    • Validation Set
    • Activation Function
• isbn_formats:
  • format: Print on Demand(PoD) ISBN, value: 978-1-7281-0324-2, isbnType: New-2005
  • format: USB ISBN, value: 978-1-7281-0322-8, isbnType: New-2005
  • format: Print ISBN, value: 978-1-7281-0321-1, isbnType: New-2005
  • format: Electronic ISBN, value: 978-1-7281-0323-5, isbnType: New-2005
• authors:
  • Author Name: Tobias Gruber
    Affiliation: Daimler AG, RD/AFU, Ulm, Germany
    Author URL: https://ieeexplore.ieee.org/author/37085998112
    ID: 37085998112
    Order: 1
    Author Affiliations:
    • Daimler AG, RD/AFU, Ulm, Germany
  • Author Name: Mariia Kokhova
    Affiliation: Institute for Photogrammetry, University of Stuttgart, Stuttgart, Germany
    Author URL: https://ieeexplore.ieee.org/author/37086543105
    ID: 37086543105
    Order: 2
    Author Affiliations:
    • Institute for Photogrammetry, University of Stuttgart, Stuttgart, Germany
  • Author Name: Werner Ritter
    Affiliation: Daimler AG, RD/AFU, Ulm, Germany
    Author URL: https://ieeexplore.ieee.org/author/37283474100
    ID: 37283474100
    Order: 3
    Author Affiliations:
    • Daimler AG, RD/AFU, Ulm, Germany
  • Author Name: Norbert Haala
    Affiliation: Institute for Photogrammetry, University of Stuttgart, Stuttgart, Germany
    Author URL: https://ieeexplore.ieee.org/author/37688184900
    ID: 37688184900
    Order: 4
    Author Affiliations:
    • Institute for Photogrammetry, University of Stuttgart, Stuttgart, Germany
  • Author Name: Klaus Dictmayer
    Affiliation: Control and Microtechnology, Ulm University, Ulm, Germany
    Author URL: https://ieeexplore.ieee.org/author/37086545990
    ID: 37086545990
    Order: 5
    Author Affiliations:
    • Control and Microtechnology, Ulm University, Ulm, Germany

Image Sensor

• sensor_type: CMOS
• resolution: 1280x720
• 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: 120 fps

Applications & Benefits

• cell_imaging: Not specified
• benefits: High contrast images even in adverse weather conditions, super-resolved depth mapping, and depth mapping aligned with pixel intensities.

Supporting Organizations

• supported_by: European Union under the H2020 ECSEL Programme as part of the DENSE project, contract number 692449.

Manuscript Details

• publication_date: 4-7 Nov. 2018

Relevancy Score

• score: 10
• missing_fields:
  • dynamic_range
  • pixel_size
  • dark_current
  • focal_length
  • aperture
  • field_of_view
  • distortion
  • sensitivity
  • noise
  • shutter_speed
  • power_consumption

Processed JSON Filename

request_ba44539a-4779-4f2b-95d6-74fb5aaab440-learning_superresolved_depth_from_active_gated_imaging.json