Using t-distributed stochastic neighbor embedding for visualization and segmentation of 3D point clouds of plants

TL;DR

This paper demonstrates that t-SNE effectively visualizes and segments 3D plant point clouds by embedding them into 2D space, facilitating plant characterization and segmentation tasks.

Contribution

It introduces the use of t-SNE for embedding 3D plant point clouds into 2D for visualization and segmentation, enabling new applications in plant phenotyping.

Findings

t-SNE effectively visualizes 3D plant structures in 2D.

Proposed simple segmentation methods work on embedded 2D data.

t-SNE embedding supports automatic phenotyping processes.

Abstract

In this work, the use of t-SNE is proposed to embed 3D point clouds of plants into 2D space for plant characterization. It is demonstrated that t-SNE operates as a practical tool to flatten and visualize a complete 3D plant model in 2D space. The perplexity parameter of t-SNE allows 2D rendering of plant structures at various organizational levels. Aside from the promise of serving as a visualization tool for plant scientists, t-SNE also provides a gateway for processing 3D point clouds of plants using their embedded counterparts in 2D. In this paper, simple methods were proposed to perform semantic segmentation and instance segmentation via grouping the embedded 2D points. The evaluation of these methods on a public 3D plant data set conveys the potential of t-SNE for enabling of 2D implementation of various steps involved in automatic 3D phenotyping pipelines.