MOT20:一种面向拥挤场景的多目标跟踪基准
计算机视觉与模式识别
2020-03-23 v1
摘要
标准化基准对绝大多数计算机视觉应用至关重要。尽管不应过度宣扬排行榜和排名表,但基准通常提供最客观的性能度量,因此是研究的重要指南。多目标跟踪基准 MOTChallenge 的推出旨在建立多目标跟踪方法的标准化评估。该挑战聚焦于多行人跟踪,因为行人在跟踪领域中被广泛研究,且精确的跟踪与检测具有很高的实际意义。自首次发布 MOT15、MOT16 和 MOT17 以来,通过引入干净的数据集和精确的框架来基准化多目标跟踪器,它们极大地推动了社区发展。在本文中,我们提出 MOT20 基准,由 8 个描绘极度拥挤困难场景的新序列组成。该基准首次发表于 2019 年计算机视觉与模式识别会议(CVPR)上的第四届 BMTT MOT Challenge 研讨会,并为评估处理极端拥挤场景的最先进(SOTA)多目标跟踪方法提供了机会。
引用
@article{arxiv.2003.09003,
title = {MOT20: A benchmark for multi object tracking in crowded scenes},
author = {Patrick Dendorfer and Hamid Rezatofighi and Anton Milan and Javen Shi and Daniel Cremers and Ian Reid and Stefan Roth and Konrad Schindler and Laura Leal-Taixé},
journal= {arXiv preprint arXiv:2003.09003},
year = {2020}
}
备注
The sequences of the new MOT20 benchmark were previously presented in the CVPR 2019 tracking challenge ( arXiv:1906.04567 ). The differences between the two challenges are: - New and corrected annotations - New sequences, as we had to crop and transform some old sequences to achieve higher quality in the annotations. - New baselines evaluations and different sets of public detections