Multiple Object Tracking from appearance by hierarchically clustering tracklets
Andreu Girbau (National Institute of Informatics),* Ferran Marques (Universitat Politecnica de Catalunya), Shin'ichi Satoh (National Institute of Informatics)
The 33rd British Machine Vision Conference
Abstract
Current approaches in Multiple Object Tracking (MOT) rely on the spatio-temporal coherence between detections combined with object appearance to match objects from consecutive frames. In this work, we explore MOT using object appearances as the main source of association between objects in a video, using spatial and temporal priors as weighting factors. We form initial tracklets by leveraging on the idea that instances of an object that are close in time should be similar in appearance, and build the final object tracks by fusing the tracklets in a hierarchical fashion. We conduct extensive experiments that show the effectiveness of our method over three different MOT benchmarks, MOT17, MOT20, and DanceTrack, being competitive in MOT17 and MOT20 and establishing state-of-the-art results in DanceTrack.Video
Citation
@inproceedings{Girbau_2022_BMVC,
author = {Andreu Girbau and Ferran Marques and Shin'ichi Satoh},
title = {Multiple Object Tracking from appearance by hierarchically clustering tracklets},
booktitle = {33rd British Machine Vision Conference 2022, {BMVC} 2022, London, UK, November 21-24, 2022},
publisher = {{BMVA} Press},
year = {2022},
url = {https://bmvc2022.mpi-inf.mpg.de/0362.pdf}
}
Copyright © 2022
The British Machine Vision Conference is organised by