Large-scene point cloud semantic segmentation method

Abstract

The invention discloses a large-scene point cloud semantic segmentation method. The method comprises the following steps: carrying out feature splicing on three-dimensional point cloud data containing feature information to obtain point cloud initial features; performing expansion graph convolution and random sampling on the point cloud initial features to obtain multi-layer intermediate features and sampling coding features; performing cross-layer context reasoning on the multi-layer intermediate features to obtain complementary context features, and splicing the complementary context features into the sampling coding features obtained in the last layer to obtain final coding features; decoding the final coding feature to obtain a decoding feature; inputting the decoding feature into a full connection layer classifier to obtain segmentation result prediction; and constructing a loss function, training and optimizing the model, and storing model parameters. According to the method, cross-layer context reasoning is used for aggregating multiple layers of contexts in the coding stage, attention fusion is adopted for feature selection in the decoding stage, information loss can be effectively made up and feature redundancy can be effectively reduced while efficiency is guaranteed, and therefore the accuracy rate is improved.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and specifically relates to an efficient and accurate semantic segmentation method for three-dimensional point clouds of large scenes using deep learning algorithms. Background technique [0002] Point cloud is one of the most basic representations of 3D scenes, which usually contains the coordinates and related features (such as color) of each point in 3D space. The task of point cloud semantic segmentation is to segment each point in the point cloud into the corresponding category through calculation and analysis. In the early days, due to the limited sensing distance, people's research mainly focused on indoor point clouds of small scenes. When processing this type of point cloud, the complete point cloud is usually divided into sub-blocks of fixed size and number of points, and on this basis, feature extraction and learning are performed on each sub-block. [0003] With the rapid de...

AI Technical Summary

Problems solved by technology

First, since a large number of points are randomly lost during feature propagation across layers, key information loss is unavoidable

The existing method of compensating for information by expanding the receptive field cannot fundamentally solve this problem because it cannot effectively perceive the loss of cross-layer information

Second, due to the loss of key information and the sparsity of large-scale point clouds, the point cloud features of the encoding layer will have noise and redundancy due to the aggregation of invalid information

Current methods usually directly stitch the encoding layer features into the decoding layer to restore the sampled point cloud, which will affect the semantic segmentation performance

Images