Point Cloud Quality Assessment

11 papers with code • 3 benchmarks • 2 datasets


A large and dense collection of points in three-dimensional space, collected by sensors such as LiDAR, is known as a point cloud. Points in the point cloud consist of geometric properties, such as three-dimensional spatial coordinates (x, y, z), and other attributes like color, reflectance, opacity, etc., represented by feature vectors. Since point clouds can directly represent the 3D world, they are widely employed in various fields, such as photogrammetry, power monitoring, architectural surveying, digital manufacturing, autonomous driving, gaming, cultural heritage reservation, and more.


Interactive point clouds typically contain millions of colored points and may possess complex attributes. To address the substantial transmission bandwidth and storage space required by point clouds, esearchers have developed various point cloud compression (PCC) techniques. However, point cloud compression may introduce significant visual distortions. In addition, deformations and distortions frequently occur during the acquisition, processing, transmission, rendering, and interaction of point clouds, all of which degrade the visual quality of the point cloud, ultimately impacting the application’s user experience. Therefore, effective methods for quantifying the quality of compressed point clouds are needed. More generally, point cloud quality assessment (PCQA) is crucial for optimizing and evaluating point cloud processing algorithms, such as encoding, denoising, and super-resolution.

Most implemented papers

No-Reference Quality Assessment for 3D Colored Point Cloud and Mesh Models

zzc-1998/NR-3DQA 5 Jul 2021

Therefore, many related studies such as point cloud quality assessment (PCQA) and mesh quality assessment (MQA) have been carried out to measure the visual quality degradations of 3D models.

Point Cloud Quality Assessment: Dataset Construction and Learning-based No-Reference Metric

lyp22/ResSCNN 22 Dec 2020

Full-reference (FR) point cloud quality assessment (PCQA) has achieved impressive progress in recent years.

Joint Geometry and Color Projection-based Point Cloud Quality Metric

AlirezaJav/Projection-based-PC-Quality-Metric 5 Aug 2021

Moreover, the proposed point cloud quality metric exploits the best performing 2D quality metrics in the literature to assess the quality of the projected images.

Perceptual Quality Assessment of Colored 3D Point Clouds

qdushl/waterloo-point-cloud-database 10 Nov 2021

In this work, we first build a large 3D point cloud database for subjective and objective quality assessment of point clouds.

No-Reference Point Cloud Quality Assessment via Domain Adaptation

qi-yangsjtu/it-pcqa CVPR 2022

We present a novel no-reference quality assessment metric, the image transferred point cloud quality assessment (IT-PCQA), for 3D point clouds.

Evaluating Point Cloud from Moving Camera Videos: A No-Reference Metric

zzc-1998/vqa_pc 30 Aug 2022

To tackle the challenge of point cloud quality assessment (PCQA), many PCQA methods have been proposed to evaluate the visual quality levels of point clouds by assessing the rendered static 2D projections.

MM-PCQA: Multi-Modal Learning for No-reference Point Cloud Quality Assessment

zzc-1998/mm-pcqa 1 Sep 2022

In specific, we split the point clouds into sub-models to represent local geometry distortions such as point shift and down-sampling.

TCDM: Transformational Complexity Based Distortion Metric for Perceptual Point Cloud Quality Assessment

zyj1318053/tcdm 10 Oct 2022

The goal of objective point cloud quality assessment (PCQA) research is to develop quantitative metrics that measure point cloud quality in a perceptually consistent manner.

No-Reference Point Cloud Quality Assessment via Weighted Patch Quality Prediction

philox12358/COPP-Net 13 May 2023

Then, we gather the features of all the patches of a point cloud for correlation analysis, to obtain the correlation weights.

GMS-3DQA: Projection-based Grid Mini-patch Sampling for 3D Model Quality Assessment

zzc-1998/gms-3dqa 9 Jun 2023

Model-based 3DQA methods extract features directly from the 3D models, which are characterized by their high degree of complexity.