preprint · arXiv (Cornell University) · 2023

Object-Centric Multiple Object Tracking

Zixu Zhao, Jiaze Wang, Max Horn, Yizhuo Ding, Tong He, Zechen Bai, Dominik Zietlow, Carl-Johann Simon-Gabriel, Bing Shuai, Zhuowen Tu, Thomas Brox, Bernt Schiele, Yanwei Fu, Francesco Locatello, Zheng Zhang, Tianjun Xiao

Cited by 1Open access

Abstract

Unsupervised object-centric learning methods allow the partitioning of scenes into entities without additional localization information and are excellent candidates for reducing the annotation burden of multiple-object tracking (MOT) pipelines. Unfortunately, they lack two key properties: objects are often split into parts and are not consistently tracked over time. In fact, state-of-the-art models achieve pixel-level accuracy and temporal consistency by relying on supervised object detection with additional ID labels for the association through time. This paper proposes a video object-centric model for MOT. It consists of an index-merge module that adapts the object-centric slots into detection outputs and an object memory module that builds complete object prototypes to handle occlusions. Benefited from object-centric learning, we only require sparse detection labels (0%-6.25%) for object localization and feature binding. Relying on our self-supervised Expectation-Maximization-inspired loss for object association, our approach requires no ID labels. Our experiments significantly narrow the gap between the existing object-centric model and the fully supervised state-of-the-art and outperform several unsupervised trackers.

Cite this paper

APA 7

Zhao, Z., Wang, J., Horn, M., Ding, Y., He, T., Bai, Z., Zietlow, D., Simon-Gabriel, C.-J., Shuai, B., Tu, Z., Brox, T., Schiele, B., Fu, Y., Locatello, F., Zhang, Z., & Xiao, T. (2023). Object-Centric Multiple Object Tracking. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2309.00233

MLA 9

Zhao, Zixu, et al. “Object-Centric Multiple Object Tracking.” arXiv (Cornell University), 2023, https://doi.org/10.48550/arxiv.2309.00233.

Chicago

Zhao, Zixu, Jiaze Wang, Max Horn, Yizhuo Ding, Tong He, Zechen Bai, Dominik Zietlow, et al. 2023. “Object-Centric Multiple Object Tracking.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2309.00233.

Harvard

Zhao, Z. et al. (2023) ‘Object-Centric Multiple Object Tracking’, arXiv (Cornell University). Available at: https://doi.org/10.48550/arxiv.2309.00233.

BibTeX

@misc{zhao2023object,
  title = {{Object-Centric Multiple Object Tracking}},
  author = {Zhao, Zixu and Wang, Jiaze and Horn, Max and Ding, Yizhuo and He, Tong and Bai, Zechen and Zietlow, Dominik and Simon-Gabriel, Carl-Johann and Shuai, Bing and Tu, Zhuowen and Brox, Thomas and Schiele, Bernt and Fu, Yanwei and Locatello, Francesco and Zhang, Zheng and Xiao, Tianjun},
  journal = {arXiv (Cornell University)},
  year = {2023},
  doi = {10.48550/arxiv.2309.00233}
}

Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.