Learning-based 3D point cloud quality assessment using a support vector regressor

Recent advances in capture technologies have increased the production of 3D content form Point Clouds (PCs). The perceived quality such data can be impacted by typical processing including acquisition, compression, transmission, visualization, etc. In this paper, we propose a learning-based method that efficiently predicts distorted PCs through set features extracted from reference PC and its degraded version. index is obtained here combining considered using Support Vector Regression (SVR) model. performance contribution each feature their combination are compared. We then discuss experimental results context state-of-the-art methods 2 publicly available datasets. also evaluate ability our to predict unknown cross-dataset evaluation. show relevance introducing learning step merge for assessment data.