Comprehensive Review of Deep Learning-Based 3D Point Cloud Completion Processing and Analysis

Point cloud completion is a generation and estimation issue derived from the partial point clouds, which plays vital role in applications of 3D computer vision. The progress deep learning (DL) has impressively improved capability robustness completion. However, quality completed clouds still needed to be further enhanced meet practical utilization. Therefore, this work aims conduct comprehensive survey on various methods, including point-based, view-based, convolution-based, graph-based, generative model-based, transformer-based approaches, etc. And summarizes comparisons among these methods provoke research insights. Besides, review sums up commonly used datasets illustrates Eventually, we also discussed possible trends promptly expanding field.