Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

BASICS: Broad quality Assessment of Static point clouds In Compression Scenarios

Ali Ak, Emin Zerman, Maurice Quach, Aladine Chetouani, Aljosa Smolic, Giuseppe Valenzise, Patrick Le Callet

TL;DR

This paper introduces BASICS, a large-scale quality assessment dataset tailored for static point clouds, and conducts a comprehensive analysis of the gathered data, benchmark existing point cloud quality assessment metrics and identify their limitations.

Abstract

Point clouds have become increasingly prevalent in representing 3D scenes within virtual environments, alongside 3D meshes. Their ease of capture has facilitated a wide array of applications on mobile devices, from smartphones to autonomous vehicles. Notably, point cloud compression has reached an advanced stage and has been standardized. However, the availability of quality assessment datasets, which are essential for developing improved objective quality metrics, remains limited. In this paper, we introduce BASICS, a large-scale quality assessment dataset tailored for static point clouds. The BASICS dataset comprises 75 unique point clouds, each compressed with four different algorithms including a learning-based method, resulting in the evaluation of nearly 1500 point clouds by 3500 unique participants. Furthermore, we conduct a comprehensive analysis of the gathered data, benchmark existing point cloud quality assessment metrics and identify their limitations. By publicly releasing the BASICS dataset, we lay the foundation for addressing these limitations and fostering the development of more precise quality metrics.

BASICS: Broad quality Assessment of Static point clouds In Compression Scenarios

TL;DR

This paper introduces BASICS, a large-scale quality assessment dataset tailored for static point clouds, and conducts a comprehensive analysis of the gathered data, benchmark existing point cloud quality assessment metrics and identify their limitations.

Abstract

Point clouds have become increasingly prevalent in representing 3D scenes within virtual environments, alongside 3D meshes. Their ease of capture has facilitated a wide array of applications on mobile devices, from smartphones to autonomous vehicles. Notably, point cloud compression has reached an advanced stage and has been standardized. However, the availability of quality assessment datasets, which are essential for developing improved objective quality metrics, remains limited. In this paper, we introduce BASICS, a large-scale quality assessment dataset tailored for static point clouds. The BASICS dataset comprises 75 unique point clouds, each compressed with four different algorithms including a learning-based method, resulting in the evaluation of nearly 1500 point clouds by 3500 unique participants. Furthermore, we conduct a comprehensive analysis of the gathered data, benchmark existing point cloud quality assessment metrics and identify their limitations. By publicly releasing the BASICS dataset, we lay the foundation for addressing these limitations and fostering the development of more precise quality metrics.
Paper Structure (23 sections, 8 figures, 5 tables)

This paper contains 23 sections, 8 figures, 5 tables.

Table of Contents

  1. Introduction
  2. The BASICS dataset
  3. Material selection
  4. Pre-processing
  5. Making 3D meshes uniform
  6. Cleaning 3D meshes
  7. Sampling point clouds
  8. Point cloud voxelization
  9. Compression
  10. Subjective quality assessment
  11. Generating Visual Stimuli
  12. Methodology
  13. Test Procedure
  14. Subjective Experiment Results
  15. Observer Screening
  16. ...and 8 more sections

Figures (8)

  • Figure 1: Sample renderings of 7 point clouds from each of the 3 semantic categories in the dataset (i.e., 7 of the total of 25 for each category).
  • Figure 2: Sample frames from the video renderings of a selected processed point cloud (PPC), showing results for each compression algorithm.
  • Figure 3: Visualization of the rendering trajectory from top and front views.
  • Figure 4: Sample screenshots from the experiment. Rendered point cloud videos were shown side-by-side (above), and each stimulus was followed by a voting screen (below).
  • Figure 5: Selected frames from the SRC and PPC video renderings of the 2 dummies used in all playlists.
  • ...and 3 more figures