# A review on deep learning techniques for 3D sensed data classification

## TL;DR

This paper reviews current deep learning methods for classifying 3D sensed data, highlighting architectures, datasets, and future research directions in a field that is less mature than 2D image understanding.

## Contribution

It provides a comprehensive overview of deep learning architectures for 3D data, including traditional methods, current approaches, datasets, and future research directions.

## Key findings

- Summarizes main deep learning approaches for 3D data classification.
- Documents datasets used for different 3D data processing methods.
- Discusses future research areas in deep learning for 3D sensed data.

## Abstract

Over the past decade deep learning has driven progress in 2D image understanding. Despite these advancements, techniques for automatic 3D sensed data understanding, such as point clouds, is comparatively immature. However, with a range of important applications from indoor robotics navigation to national scale remote sensing there is a high demand for algorithms that can learn to automatically understand and classify 3D sensed data. In this paper we review the current state-of-the-art deep learning architectures for processing unstructured Euclidean data. We begin by addressing the background concepts and traditional methodologies. We review the current main approaches including; RGB-D, multi-view, volumetric and fully end-to-end architecture designs. Datasets for each category are documented and explained. Finally, we give a detailed discussion about the future of deep learning for 3D sensed data, using literature to justify the areas where future research would be most valuable.