A comparison between single-stage and two-stage 3D tracking algorithms for greenhouse robotics

TL;DR

This paper compares single-stage and two-stage 3D multi-object tracking algorithms in greenhouse robotics, demonstrating that single-stage methods outperform two-stage ones in complex, occlusion-heavy scenarios for improved object tracking accuracy.

Contribution

The study provides a comparative analysis of 3D-SORT and MOT-DETR algorithms in real greenhouse conditions, highlighting the advantages of single-stage methods in occlusion scenarios.

Findings

Single-stage MOT-DETR outperforms two-stage 3D-SORT in accuracy.

Single-stage method excels in occlusion-heavy sequences.

Complex sequences benefit more from single-stage tracking.

Abstract

With the current demand for automation in the agro-food industry, accurately detecting and localizing relevant objects in 3D is essential for successful robotic operations. However, this is a challenge due the presence of occlusions. Multi-view perception approaches allow robots to overcome occlusions, but a tracking component is needed to associate the objects detected by the robot over multiple viewpoints. Multi-object tracking (MOT) algorithms can be categorized between two-stage and single-stage methods. Two-stage methods tend to be simpler to adapt and implement to custom applications, while single-stage methods present a more complex end-to-end tracking method that can yield better results in occluded situations at the cost of more training data. The potential advantages of single-stage methods over two-stage methods depends on the complexity of the sequence of viewpoints that a robot needs to process. In this work, we compare a 3D two-stage MOT algorithm, 3D-SORT, against a 3D single-stage MOT algorithm, MOT-DETR, in three different types of sequences with varying levels of complexity. The sequences represent simpler and more complex motions that a robot arm can perform in a tomato greenhouse. Our experiments in a tomato greenhouse show that the single-stage algorithm consistently yields better tracking accuracy, especially in the more challenging sequences where objects are fully occluded or non-visible during several viewpoints.