Multi-Scale Histogram-Based Probabilistic Deep Neural Network for Super-Resolution 3D LiDAR Imaging

Abstract

LiDAR (Light Detection and Ranging) imaging based on SPAD (Single-Photon Avalanche Diode) technology suffers from severe area penalty for large on-chip histogram peak detection circuits required by the high precision of measured depth values. In this work, a probabilistic estimation-based super-resolution neural network for SPAD imaging that firstly uses temporal multi-scale histograms as inputs is proposed. To reduce the area and cost of on-chip histogram computation, only part of the histogram hardware for calculating the reflected photons is implemented on a chip. On account of the distribution rule of returned photons, a probabilistic encoder as a part of the network is first proposed to solve the depth estimation problem of SPADs. By jointly using this neural network with a super-resolution network, 16× up-sampling depth estimation is realized using 32 × 32 multi-scale histogram outputs. Finally, the effectiveness of this neural network was verified in the laboratory with a 32 × 32 SPAD sensor system.

Usage of a Conventional Device with LiDAR Implementation for Mesh Model Creation

“…As for the pragmatic, technical approach, it is possible to compare the dimensions of the object, the identity of the edges, the shape of the object, and so on. For this work, the method of comparing the coordinates of uniquely determined points on an object with irregular fitting was chosen [41][42][43][44]. After the detailed points were fitted on the object, these points were aligned using the conventional spatial polar method, and their spatial coordinates were determined in the relative coordinate system.…”