Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

PCFEx: Point Cloud Feature Extraction for Graph Neural Networks

Abdullah Al Masud, Shi Xintong, Mondher Bouazizi, Ohtsuki Tomoaki

TL;DR

This work proposes novel point cloud feature extraction (PCFEx) techniques to capture meaningful information at the point, edge, and graph levels of the point cloud by considering point cloud as a graph and introduces a GNN architecture designed to efficiently process these features.

Abstract

Graph neural networks (GNNs) have gained significant attention for their effectiveness across various domains. This study focuses on applying GNN to process 3D point cloud data for human pose estimation (HPE) and human activity recognition (HAR). We propose novel point cloud feature extraction (PCFEx) techniques to capture meaningful information at the point, edge, and graph levels of the point cloud by considering point cloud as a graph. Moreover, we introduce a GNN architecture designed to efficiently process these features. Our approach is evaluated on four most popular publicly available millimeter wave radar datasets, three for HPE and one for HAR. The results show substantial improvements, with significantly reduced errors in all three HPE benchmarks, and an overall accuracy of 98.8% in mmWave-based HAR, outperforming the existing state of the art models. This work demonstrates the great potential of feature extraction incorporated with GNN modeling approach to enhance the precision of point cloud processing.

PCFEx: Point Cloud Feature Extraction for Graph Neural Networks

TL;DR

This work proposes novel point cloud feature extraction (PCFEx) techniques to capture meaningful information at the point, edge, and graph levels of the point cloud by considering point cloud as a graph and introduces a GNN architecture designed to efficiently process these features.

Abstract

Graph neural networks (GNNs) have gained significant attention for their effectiveness across various domains. This study focuses on applying GNN to process 3D point cloud data for human pose estimation (HPE) and human activity recognition (HAR). We propose novel point cloud feature extraction (PCFEx) techniques to capture meaningful information at the point, edge, and graph levels of the point cloud by considering point cloud as a graph. Moreover, we introduce a GNN architecture designed to efficiently process these features. Our approach is evaluated on four most popular publicly available millimeter wave radar datasets, three for HPE and one for HAR. The results show substantial improvements, with significantly reduced errors in all three HPE benchmarks, and an overall accuracy of 98.8% in mmWave-based HAR, outperforming the existing state of the art models. This work demonstrates the great potential of feature extraction incorporated with GNN modeling approach to enhance the precision of point cloud processing.
Paper Structure (39 sections, 9 equations, 6 figures, 9 tables)

This paper contains 39 sections, 9 equations, 6 figures, 9 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Convolutional neural networks (CNN)
  4. Multi layer perceptrons (MLP)
  5. Transformers
  6. Graph neural networks (GNN)
  7. Proposed Method
  8. Definition
  9. Data Processing
  10. Frame Fusion
  11. Node Downsampling
  12. Node Feature Extraction
  13. Edge Feature Extraction
  14. Frame Feature Extraction
  15. GNN Model Architecture
  16. ...and 24 more sections

Figures (6)

  • Figure 1: (a) point cloud graph formation of frame $i$; (b) data-processing pipeline.
  • Figure 2: Node feature map (right) of a 3D point cloud (left). Points 7, 12, 27, 32 show large centroid distances; point 36 may be an outlier due to high neighbor distance, standard deviation.
  • Figure 3: Node feature extraction.
  • Figure 4: Frame feature extraction.
  • Figure 5: Proposed graph neural network. This architecture calculates the frame representation feature vector.
  • ...and 1 more figures