ROV6D: 6D Pose Estimation Benchmark Dataset for Underwater Remotely Operated Vehicles

Accurately localization between multi-robots is crucial for many underwater applications, such as tracking, convoying and subsea intervention tasks. 6D pose estimation is a fundamental task that enables precise object localization in 3D space with full six degrees of freedom. However, one critical challenge is the lack of available large-scale datasets due to the unbearable cost of labelled data collection. To overcome this difficulty, we propose a benchmark dataset, ROV6D, for 6D pose estimation of remotely operated vehicles (ROVs). The training subset consists of a large number of synthetic images with 6D pose ground truth for ROVs. These synthetic images are generated using BlenderProc and further rendered with the underwater neural rendering (UWNR) strategy to enhance their realism. The testing subsets cover different real-world scenarios, including the Pool subset and Maoming subset, focusing on challenging cases that involve partial occlusion and low visibility. Diverse recent methods are evaluated on the constructed dataset. The results show that methods based on dense coordinates currently perform best, outperforming both the keypoint-based method and the refinement-based method. Our dataset will be made publicly available soon.

I. Introduction

Estimating the 6D pose, i.e., the 3D rotation and 3D translation, is essential for underwater localization, navigation and cooperative operations between unmanned underwater vehicles (UUVs), such as remotely operated vehicles (ROVs) and autonomous underwater vehicles (AUVs). The vast majority of previous work about multi-robot relative localization focus on terrestrial autonomous driving, on-orbit servicing and unmanned aerial vehicle [1], [2], [3], [4], [5]. Although researches on 6D pose estimation are proliferating in recent years, some important environments have not received significant attention thus far, such as underwater.