A Two-Stage Minimum Cost Multicut Approach to Self-Supervised Multiple Person Tracking

Kalun Ho, Amirhossein Kardoost, Franz-Josef Pfreundt, Janis Keuper, Margret Keuper; Proceedings of the Asian Conference on Computer Vision (ACCV), 2020

Abstract


Multiple Object Tracking (MOT) is a long-standing task in computer vision. Current approaches based on the tracking by detection paradigm either require some sort of domain knowledge or supervision to associate data correctly into tracks. In this work, we present a self-supervised multiple object tracking approach based on visual features and minimum cost lifted multicuts. Our method is based on straight-forward spatio-temporal cues that can be extracted from neighboring frames in an image sequences without supervision. Clustering based on these cues enables us to learn the required appearance invariances for the tracking task at hand and train an AutoEncoder to generate suitable latent representations. Thus, the resulting latent representations can serve as robust appearance cues for tracking even over large temporal distances where no reliable spatio-temporal features can be extracted. We show that, despite being trained without using the provided annotations, our model provides competitive results on the challenging MOT Benchmark for pedestrian tracking.

Related Material


[pdf] [supp]
[bibtex]
@InProceedings{Ho_2020_ACCV, author = {Ho, Kalun and Kardoost, Amirhossein and Pfreundt, Franz-Josef and Keuper, Janis and Keuper, Margret}, title = {A Two-Stage Minimum Cost Multicut Approach to Self-Supervised Multiple Person Tracking}, booktitle = {Proceedings of the Asian Conference on Computer Vision (ACCV)}, month = {November}, year = {2020} }