Home

Awesome

Quasi-Dense Tracking

PWC PWC

News

We released a new version of our paper with new benchmark results setting a new SOTA on BDD100K!

QDTrack: Quasi-Dense Similarity Learning for Appearance-Only Multiple Object Tracking

This is the offical implementation of paper Quasi-Dense Similarity Learning for Multiple Object Tracking.

We present a trailer that consists of method illustrations and tracking visualizations. Our project website contains more information: vis.xyz/pub/qdtrack.

If you have any questions, please go to Discussions.

Abstract

Similarity learning has been recognized as a crucial step for object tracking. However, existing multiple object tracking methods only use sparse ground truth matching as the training objective, while ignoring the majority of the informative regions on the images. In this paper, we present Quasi-Dense Similarity Learning, which densely samples hundreds of region proposals on a pair of images for contrastive learning. We can naturally combine this similarity learning with existing detection methods to build Quasi-Dense Tracking (QDTrack) without turning to displacement regression or motion priors. We also find that the resulting distinctive feature space admits a simple nearest neighbor search at the inference time. Despite its simplicity, QDTrack outperforms all existing methods on MOT, BDD100K, Waymo, and TAO tracking benchmarks. It achieves 68.7 MOTA at 20.3 FPS on MOT17 without using external training data. Compared to methods with similar detectors, it boosts almost 10 points of MOTA and significantly decreases the number of ID switches on BDD100K and Waymo datasets.

Quasi-dense matching

<img src="figures/teaser.png" width="600">

Main results

Without bells and whistles, our method outperforms the states of the art on MOT, BDD100K, Waymo, and TAO benchmarks with ResNet-50 as the base model.

BDD100K test set

mMOTAmIDF1ID Sw.
35.552.310790

MOT

DatasetMOTAIDF1ID Sw.MTML
MOT1669.867.11097316150
MOT1768.766.33378957516

Waymo validation set

CategoryMOTAIDF1ID Sw.
Vehicle55.666.224309
Pedestrian50.358.46347
Cyclist26.245.756
All44.056.830712

TAO

SplitAP50AP75AP
val16.15.07.0
test12.44.55.2

Installation

Please refer to INSTALL.md for installation instructions.

Usages

Please refer to GET_STARTED.md for dataset preparation and running instructions.

Trained models for testing

More implementations / models on the following benchmarks will be released later

Waymo models won't be available publicly due to the dataset license constraints.

Citation

@article{qdtrack,
  title={QDTrack: Quasi-Dense Similarity Learning for Appearance-Only Multiple Object Tracking},
  author={Fischer, Tobias and Pang, Jiangmiao and Huang, Thomas E and Qiu, Linlu and Chen, Haofeng and Darrell, Trevor and Yu, Fisher},
  journal={arXiv preprint arXiv:2210.06984},
  year={2022}
}

@InProceedings{qdtrack_conf,
  title = {Quasi-Dense Similarity Learning for Multiple Object Tracking},
  author = {Pang, Jiangmiao and Qiu, Linlu and Li, Xia and Chen, Haofeng and Li, Qi and Darrell, Trevor and Yu, Fisher},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  month = {June},
  year = {2021}
}