Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Full-reference Point Cloud Quality Assessment Using Spectral Graph Wavelets

Ryosuke Watanabe, Keisuke Nonaka, Eduardo Pavez, Tatsuya Kobayashi, Antonio Ortega

TL;DR

A full-reference PCQA method utilizing spectral graph wavelets (SGWs) utilizing spectral graph wavelets (SGWs) and novel SGW-based PCQA metrics that compare SGW coefficients of coordinate and color signals between reference and distorted point clouds are introduced.

Abstract

Point clouds in 3D applications frequently experience quality degradation during processing, e.g., scanning and compression. Reliable point cloud quality assessment (PCQA) is important for developing compression algorithms with good bitrate-quality trade-offs and techniques for quality improvement (e.g., denoising). This paper introduces a full-reference (FR) PCQA method utilizing spectral graph wavelets (SGWs). First, we propose novel SGW-based PCQA metrics that compare SGW coefficients of coordinate and color signals between reference and distorted point clouds. Second, we achieve accurate PCQA by integrating several conventional FR metrics and our SGW-based metrics using support vector regression. To our knowledge, this is the first study to introduce SGWs for PCQA. Experimental results demonstrate the proposed PCQA metric is more accurately correlated with subjective quality scores compared to conventional PCQA metrics.

Full-reference Point Cloud Quality Assessment Using Spectral Graph Wavelets

TL;DR

A full-reference PCQA method utilizing spectral graph wavelets (SGWs) utilizing spectral graph wavelets (SGWs) and novel SGW-based PCQA metrics that compare SGW coefficients of coordinate and color signals between reference and distorted point clouds are introduced.

Abstract

Point clouds in 3D applications frequently experience quality degradation during processing, e.g., scanning and compression. Reliable point cloud quality assessment (PCQA) is important for developing compression algorithms with good bitrate-quality trade-offs and techniques for quality improvement (e.g., denoising). This paper introduces a full-reference (FR) PCQA method utilizing spectral graph wavelets (SGWs). First, we propose novel SGW-based PCQA metrics that compare SGW coefficients of coordinate and color signals between reference and distorted point clouds. Second, we achieve accurate PCQA by integrating several conventional FR metrics and our SGW-based metrics using support vector regression. To our knowledge, this is the first study to introduce SGWs for PCQA. Experimental results demonstrate the proposed PCQA metric is more accurately correlated with subjective quality scores compared to conventional PCQA metrics.
Paper Structure (20 sections, 9 equations, 4 figures, 2 tables)

This paper contains 20 sections, 9 equations, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related work
  3. 3D-to-2D projection-based methods
  4. 3D model-based methods
  5. Learning-based methods
  6. Preliminaries
  7. Graph and point cloud notations
  8. Spectral graph wavelet transform (SGWT)
  9. Proposed method
  10. Overview of the proposed method
  11. Calculation of spectral graph wavelets
  12. SGW-based FR-PCQA metrics
  13. Score integration by support vector regression (SVR)
  14. Extension of (+)
  15. EXPERIMENTS
  16. ...and 5 more sections

Figures (4)

  • Figure 1: Calculation of spectral graph wavelet transforms (SGWTs) for a point cloud.
  • Figure 2: Overall calculation flow of the proposed metrics using SGWs.
  • Figure 3: Changes of SGWs through distortions caused by (a) Gaussian noise (geometry), (b) downsampling, (c) Gaussian noise (color), and (d) compression error (G-PCC RAHT MPEGPCC).
  • Figure 4: SROCCs of each score, (a) geometry-based score ($S^G_m$), and (b) color-based score ($S^C_m$), with the BASICS dataset (test set).