Point cloud data is a large collection of high dimensional 3D points with 3D coordinates and attributes, which has been one of the mainstream representations for emerging 3D applications, such as virtual reality, autonomous vehicles and robotics. Due to the large-scale unstructured high-dimensional nature of point clouds, point cloud processing, transmitting and analysing has been challenging issues in multimedia signal processing and communication. Deep learning is a powerful tool to learn statistical knowledge from massive data. Advances in artificial intelligence, especially deep learning models are offering new opportunities for point cloud processing, compression and analysis. This special issue aims at promoting cutting-edge research on deep learning-based point cloud processing, including object detection, segmentation, registration, compression, and visual quality assessment.

In this letter, the colored point cloud quality assessment in Augmented Reality (AR) environment was fully studied through subjective test. Firstly, we present a point cloud dataset, named Point Cloud Quality Dataset-AR (PCQD-AR), including ten reference point clouds and their 90 distorted versions, which were encoded by the reference software of Video-based Point Cloud Compression (V-PCC) under different pairs of geometry and texture quantization parameters. Then, the impact of geometry and texture distortions on perceived quality of point clouds in the AR environment was discussed in detail. Moreover, we evaluate the performance of existing objective point cloud quality assessment metrics on the proposed dataset. The subjective dataset including the values of Mean Opinion Score (MOS) will be released after acceptance.