Occupancy-Assisted Attribute Artifact Reduction for Video-Based Point Cloud Compression

Linyao Gao,Zhu Li,Lizhi Hou,Yiling Xu,Jun Sun

DOI: https://doi.org/10.1109/tbc.2024.3353568

IF: 4.5

2024-01-01

IEEE Transactions on Broadcasting

Abstract:Video-based point cloud compression (V-PCC) has achieved remarkable compression efficiency, which converts point clouds into videos and leverages video codecs for coding. For lossy compression, the undesirable artifacts of attribute images always degrade the point clouds attribute reconstruction quality. In this paper, we propose an Occupancy-assisted Compression Artifact Removal Network (OCARNet) to remove the distortions of V-PCC decoded attribute images for high-quality point cloud attribute reconstruction. Specifically, the occupancy information is fed into network as a prior knowledge to provide more spatial and structural information and to assist in eliminating the distortions of the texture regions. To aggregate the occupancy information effectively, we design a multi-level feature fusion framework with Channel-Spatial Attention based Residual Blocks (CSARB), where the short and long residual connections are jointly employed to capture the local context and long-range dependency. Besides, we propose a Masked Mean Square Error (MMSE) loss function based on the occupancy information to train our proposed network to focus on estimating the attribute artifacts of the occupied regions. To the best of our knowledge, this is the first learning-based attribute artifact removal method for V-PCC. Experimental results demonstrate that our framework outperforms existing state-of-the-art methods and shows the effectiveness on both objective and subjective quality comparisons.