Abstract
Single-photon avalanche diodes (SPAD) are powerful sensors for 3D light detection and ranging (LiDAR) in low light scenarios due to their single-photon sensitivity. However, accurately retrieving ranging information from noisy time-of-arrival (ToA) point clouds remains a challenge. This paper proposes a photon-efficient, non-fusion neural network architecture that can directly reconstruct high-fidelity depth images from ToA data without relying on other guiding images. Besides, the neural network architecture was compressed via a low-bit quantization scheme so that it is suitable to be implemented on embedded hardware platforms. The proposed quantized neural network architecture achieves superior reconstruction accuracy and fewer parameters than previously reported networks.
Highlights
Depth imaging has been an essential tool in various applications, such as autonomous vehicles [1], vision-guided robotic systems [2], and augmented reality applications [3]
We proposed a photon-efficient 3D convolutional neural network (CNN) architecture without using sensor-fusion strategies
We evaluated our compressed model over various signal-background ratio (SBR) levels in terms of extensive evaluation metrics [25] for synthetic and captured data [20] with a high spatial resolution
Summary
Depth imaging has been an essential tool in various applications, such as autonomous vehicles [1], vision-guided robotic systems [2], and augmented reality applications [3]. Many strategies have been proposed to obtain depth information from captured intensity images. The stereovision technique mimicking human vision systems [5] is popular. It uses the triangulation principle to understand spatial information. The sensing accuracy of the stereovision approach deteriorates when it works in dark conditions or performs long-distance measurements, whereas LiDAR can overcome the limitations. Unlike Radar systems [6] that use radio waves to measure the time-of-flight (ToF) between transmitted and reflected signals, LiDAR systems adopt pulsed light with a much shorter wavelength to detect an object’s range. LiDAR can obtain more accurate spatial information than Radar for seeing objects at a longer distance [7]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
