Three-dimensional visual inspection method, system and device based on shape attention mechanism

Abstract

The invention belongs to the field of computer deep reinforcement learning and pattern recognition, particularly relates to a three-dimensional visual inspection method, system and device based on a shape attention mechanism, and aims to solve the problems that the precision of a single-stage detector is lower than that of a two-stage detector, and the two-stage detector consumes a large amount oftime and is not suitable for a real-time system. The method comprises the steps of representing point cloud data through three-dimensional grid voxels; extracting features and encoding a spatial sparse feature map; after projecting to a top view, extracting different scale features; combining the features by adopting a deconvolution layer; extracting a shape attention feature map through the attention weight and the convolutional coding layer; and obtaining a target category, a target position, a target size and a target direction through the target classification network and the regression positioning network. According to the method, a sampling strategy based on distance constraint and an attention mechanism based on shape prior are used, instability caused by uneven data distribution is relieved, the problem that a single-stage detector lacks shape prior is solved, and the method is high in precision, short in consumed time, high in real-time performance and good in robustness.

Description

technical field [0001] The invention belongs to the fields of deep reinforcement learning, computer vision, pattern recognition and machine learning, and specifically relates to a three-dimensional visual detection method, system and device based on a shape attention mechanism. Background technique [0002] 3D object detectors need to output reliable spatial and semantic information, namely 3D position, orientation, occupied volume and category. Compared with 2D object detection, 3D objects provide more detailed information, but modeling is more difficult. 3D object detection generally uses distance sensors, such as lidar, TOF cameras, stereo cameras, etc., to predict more meaningful and accurate results. 3D object detection has become a key technology in areas such as self-driving cars, UVA, and robotics. Most accurate 3D object detection algorithms in traffic scenes are based on radar sensors, which have become the basic sensor for outdoor scene perception. Object perce...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the above problems in the prior art, that is, the accuracy of the single-stage three-dimensional object detector is lower than that of the two-stage detector, and the two-stage detector is time-consuming and not suitable for real-time systems, the present invention provides a shape-based A three-dimensional visual detection method of attention mechanism, the three-dimensional target detection method includes:

Images