Detection and tracking of objects in underwater video

TLDR

ROVs routinely capture hours of video daily, but manual processing is a bottleneck, and low‑contrast translucent targets are hard to detect because of variable lighting and marine‑snow noise. The study aims to develop and evaluate methods that overcome detection challenges in ROV video. The system uses a selective‑attention algorithm to pre‑select salient targets for track initiation, thereby simplifying multi‑target tracking and the assignment problem. The automated system successfully detects and tracks objects of interest in ROV video, improving efficiency for human annotators.

Abstract

For oceanographic research, remotely operated underwater vehicles (ROVs) routinely record several hours of video material each day. Manual processing of such large amounts of video has become a major bottleneck for scientific research based on this data. We have developed an automated system that detects and tracks objects that are of potential interest for human video annotators. By pre-selecting salient targets for track initiation using a selective attention algorithm, we reduce the complexity of multi-target tracking, in particular of the assignment problem. Detection of low-contrast translucent targets is difficult due to variable lighting conditions and the presence of ubiquitous noise from high-contrast organic debris ("marine snow") particles. We describe the methods we developed to overcome these issues and report our results of processing ROV video data.