3D Point Cloud Descriptors in Hand-crafted and Deep Learning Age: State-of-the-Art

TL;DR

This paper provides a comprehensive review of 3D point cloud descriptors, comparing traditional hand-crafted methods with modern deep learning approaches, highlighting their strengths, limitations, and future research directions.

Contribution

It offers an in-depth classification and analysis of existing 3D point cloud descriptors, emphasizing the evolution from hand-crafted to deep learning-based methods.

Findings

Hand-crafted descriptors are limited in robustness and discriminability.

Deep learning approaches improve feature extraction but require large datasets.

Future research should focus on hybrid methods and real-time applications.

Abstract

The introduction of inexpensive 3D data acquisition devices has promisingly facilitated the wide availability and popularity of 3D point cloud, which attracts more attention to the effective extraction of novel 3D point cloud descriptors for accuracy of the efficiency of 3D computer vision tasks in recent years. However, how to develop discriminative and robust feature descriptors from 3D point cloud remains a challenging task due to their intrinsic characteristics. In this paper, we give a comprehensively insightful investigation of the existing 3D point cloud descriptors. These methods can principally be divided into two categories according to the advancement of descriptors: hand-crafted based and deep learning-based apporaches, which will be further discussed from the perspective of elaborate classification, their advantages, and limitations. Finally, we present the future research direction of the extraction of 3D point cloud descriptors.