BrackishMOT

This is a MOT expansion to the [Brackish Dataset](https://www.kaggle.com/datasets/aalborguniversity/brackish-dataset) which include annotations that follows the [MOTChallenge standard](https://motchallenge.net/) and synthetic sequences that can be used for training. An additional nine real sequences containing the *small fish* class have been added, which are not part of the original Brackish Dataset. More information about **BrackishMOT** can be found in the paper **[BrackishMOT: The Brackish Multi-Object Tracking Dataset](https://arxiv.org/abs/2302.10645)** (accepted at SCIA 2023). ### Abstract There exist no publicly available annotated underwater multi-object tracking (MOT) datasets captured in turbid environments. To remedy this we propose the BrackishMOT dataset with focus on tracking schools of small fish, which is a notoriously difficult MOT task. BrackishMOT consists of 98 sequences captured in the wild. Alongside the novel dataset, we present baseline results by training a state-of-the-art tracker. Additionally, we propose a framework for creating synthetic sequences in order to expand the dataset. The framework consists of animated fish models and realistic underwater environments. We analyse the effects of including synthetic data during training and show that a combination of real and synthetic underwater training data can enhance tracking performance. Project page: https://www.vap.aau.dk/brackishmot  ### Citation ``` @InProceedings{Pedersen_2023, author = {Pedersen, Malte and Lehotský, Daniel and Nikolov, Ivan and Moeslund, Thomas B.}, doi = {10.48550/ARXIV.2302.10645}, title = {BrackishMOT: The Brackish Multi-Object Tracking Dataset}, publisher={arXiv}, year={2023} } ```本数据集是对[咸水数据集](https://www.kaggle.com/datasets/aalborguniversity/brackish-dataset)的MOT（多目标跟踪）扩展，其中包含遵循[MOTChallenge标准](https://motchallenge.net/)的标注以及用于训练的合成序列。新增了九个包含*小鱼*类别的真实序列，这些序列并非原始咸水数据集的一部分。 更多关于**BrackishMOT**的信息可在论文**[BrackishMOT: 咸水多目标跟踪数据集](https://arxiv.org/abs/2302.10645)**（已被SCIA 2023接受）中找到。 ### 摘要 目前尚无公开可用的在浑浊环境中捕获的标注水下多目标跟踪（MOT）数据集。为解决此问题，我们提出了BrackishMOT数据集，重点关注小鱼群跟踪，这是一项众所周知极具挑战性的MOT任务。BrackishMOT包含98个在野外捕获的序列。在新的数据集的基础上，我们通过训练最先进的跟踪器展示了基线结果。此外，我们提出了一种创建合成序列的框架，以扩展数据集。该框架由动画鱼模型和逼真的水下环境组成。我们分析了在训练过程中包含合成数据的效果，并表明真实和合成水下训练数据的结合可以提升跟踪性能。 项目页面：https://www.vap.aau.dk/brackishmot  ### 引用 @InProceedings{Pedersen_2023, author = {Pedersen, Malte and Lehotský, Daniel and Nikolov, Ivan and Moeslund, Thomas B.}, doi = {10.48550/ARXIV.2302.10645}, title = {BrackishMOT: The Brackish Multi-Object Tracking Dataset}, publisher={arXiv}, year={2023} }

搜集汇总

背景与挑战

背景概述

BrackishMOT是一个水下浑浊环境中的多目标追踪数据集，专注于追踪小鱼群，包含98个真实和合成序列。该数据集填补了该领域公开数据集的空白，并提供合成数据扩展框架以提升追踪性能。