Blind Quality Assessment of Dense 3D Point Clouds with Structure Guided Resampling

What is it about?

Objective quality assessment of 3D point clouds is essential for the development of immersive multimedia systems in real-world applications. Despite the success of perceptual quality evaluation for 2D images and videos, blind/no-reference metrics are still scarce for 3D point clouds with large-scale irregularly distributed 3D points. Therefore, we propose an objective point cloud quality index with Structure Guided Resampling (SGR) to automatically evaluate the perceptually visual quality of dense 3D point clouds. The proposed SGR is a general-purpose blind quality assessment method without the assistance of any reference information.

Featured Image

Photo by Ion (Ivan) Sipilov on Unsplash

Photo by Ion (Ivan) Sipilov on Unsplash

Why is it important?

With the modern advances in 3D data capture devices and rendering technologies, popular 3D point clouds have become one of the most important multimedia representations for providing 6 degrees of freedom. As defined, a 3D point cloud contains a huge number of scattered points with certain attributes. Each 3D point owns geometry and attribute information, which can represent actual objects or environments more visually compared with traditional images and videos. Due to such an abundant pattern, there have emerged lots of real-world applications for 3D point clouds, including VR/AR/XR, automatic driving, and scene understanding. A variety of distortions would be inevitably introduced during the processing chain of multimedia communication systems for point clouds. Thus, developing efficient objective quality metrics that can automatically evaluate the perceptual quality of point clouds is important.

Perspectives