Deep Learning Multi-View Fusion for Blind 3D Point Cloud Quality Assessment
Résumé
In recent years, 3D point clouds (3DPC) have experienced rapid growth in various
fields of computer vision, leading to an increased demand for efficient approaches to
automatically assess the quality of 3D point clouds. In this work, we propose a deep
learning-based method for No-Reference (Blind) Point Cloud Quality Assessment
(NR-PCQA) that aims to automatically predict the perceived visual quality of the
3DPC without relying on a reference content. We evaluate the performance of our
model on two benchmark databases: SJTU and WPC. Experimental results show
that our model achieves good performance compared to state-of-the-art methods.