A three-dimensional target detection method and system for mechanical assemblies based on CAD model domain point cloud data

Abstract

The present invention belongs to the technical field of three-dimensional target detection, and specifically discloses a three-dimensional target detection method and system for mechanical assemblies based on CAD model domain point cloud data, which includes: obtaining a model domain point cloud and corresponding labels based on the CAD three-dimensional model of the mechanical assembly, the labels reflecting the category and position of the parts; training the detector and domain discriminator through the model domain and real domain point cloud data, so as to maximize the domain discriminator error and minimize the detector error as the loss function; the detector includes a feature extractor and a label predictor, the feature extractor extracts point cloud features, the domain discriminator discriminates whether the features come from the model domain or the real domain, and the label predictor outputs the corresponding label according to the model features. The real domain point cloud is input into the trained detector, and the corresponding label is output. The present invention can realize the accurate identification and positioning of parts of the assembled assembly.

Images