Saved in:
Bibliographic Details
Main Authors: Karaev, Nikita, Rocco, Ignacio, Graham, Benjamin, Neverova, Natalia, Vedaldi, Andrea, Rupprecht, Christian
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2307.07635
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910625638121472
author Karaev, Nikita
Rocco, Ignacio
Graham, Benjamin
Neverova, Natalia
Vedaldi, Andrea
Rupprecht, Christian
author_facet Karaev, Nikita
Rocco, Ignacio
Graham, Benjamin
Neverova, Natalia
Vedaldi, Andrea
Rupprecht, Christian
contents We introduce CoTracker, a transformer-based model that tracks a large number of 2D points in long video sequences. Differently from most existing approaches that track points independently, CoTracker tracks them jointly, accounting for their dependencies. We show that joint tracking significantly improves tracking accuracy and robustness, and allows CoTracker to track occluded points and points outside of the camera view. We also introduce several innovations for this class of trackers, including using token proxies that significantly improve memory efficiency and allow CoTracker to track 70k points jointly and simultaneously at inference on a single GPU. CoTracker is an online algorithm that operates causally on short windows. However, it is trained utilizing unrolled windows as a recurrent network, maintaining tracks for long periods of time even when points are occluded or leave the field of view. Quantitatively, CoTracker substantially outperforms prior trackers on standard point-tracking benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2307_07635
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle CoTracker: It is Better to Track Together
Karaev, Nikita
Rocco, Ignacio
Graham, Benjamin
Neverova, Natalia
Vedaldi, Andrea
Rupprecht, Christian
Computer Vision and Pattern Recognition
We introduce CoTracker, a transformer-based model that tracks a large number of 2D points in long video sequences. Differently from most existing approaches that track points independently, CoTracker tracks them jointly, accounting for their dependencies. We show that joint tracking significantly improves tracking accuracy and robustness, and allows CoTracker to track occluded points and points outside of the camera view. We also introduce several innovations for this class of trackers, including using token proxies that significantly improve memory efficiency and allow CoTracker to track 70k points jointly and simultaneously at inference on a single GPU. CoTracker is an online algorithm that operates causally on short windows. However, it is trained utilizing unrolled windows as a recurrent network, maintaining tracks for long periods of time even when points are occluded or leave the field of view. Quantitatively, CoTracker substantially outperforms prior trackers on standard point-tracking benchmarks.
title CoTracker: It is Better to Track Together
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2307.07635