Multi Target proposal MixFormer and KF-based liner occlusion handling

This is a trimmed-down version of MixFormer to continue on proposed implementation by adding the following features:

• Isolate Mixformer Convmae Online model so it is easy to integrate into other applications ✅
• Add an auxiliary Linear motion model Kalman filter to avoid id switch for a short period ✅
• Implement reassociation methodology and heuristic for kalman update
• Implement center point based multi detection head
• Add an auxiliary EKF or PSO Based Object Tracking module to maintain tracklet for the original target

MixFormer (Base REPO)

MixFormer

✨ Strong performance

Tracker	VOT2020 (EAO)	LaSOT (NP)	GOT-10K (AO)	TrackingNet (NP)
MixViT-L (ConvMAE)	0.567	82.8	-	90.3
MixViT-L	0.584	82.2	75.7	90.2
MixCvT	0.555	79.9	70.7	88.9
ToMP101* (CVPR2022)	-	79.2	-	86.4
SBT-large* (CVPR2022)	0.529	-	70.4	-
SwinTrack* (Arxiv2021)	-	78.6	69.4	88.2
Sim-L/14* (Arxiv2022)	-	79.7	69.8	87.4
STARK (ICCV2021)	0.505	77.0	68.8	86.9
KeepTrack (ICCV2021)	-	77.2	-	-
TransT (CVPR2021)	0.495	73.8	67.1	86.7
TrDiMP (CVPR2021)	-	-	67.1	83.3
Siam R-CNN (CVPR2020)	-	72.2	64.9	85.4
TREG (Arxiv2021)	-	74.1	66.8	83.8

Install the environment

Use the Anaconda

conda create -n mixformer python=3.6
conda activate mixformer
bash install_pytorch17.sh

Test and evaluate MixFormer on Your own video

Only Isolated model (change vid path to your own video)

python mixformer_convmae_online.py 

Model with Linear KF

from KF_tracker_wrapper import TrackingModel_2D

tracker = TrackingModel_2D()
vid_path = 'path/to/your/video'
tracker.track(vid_path)

Model Zoo and raw results

The trained models and the raw tracking results are provided in the [Models and Raw results] (Google Driver) or [Models and Raw results] (Baidu Driver: hmuv).

Contact

Fawad Abbas: [email protected]

Yutao Cui(Orignal Autohr): [email protected]

Acknowledgments

• Thanks for MCG-NJU for opensourcing their implementation

✏️ Citation

If you think this project is helpful, please feel free to leave a star⭐️ and cite author's paper:

@inproceedings{cui2022mixformer,
  title={Mixformer: End-to-end tracking with iterative mixed attention},
  author={Cui, Yutao and Jiang, Cheng and Wang, Limin and Wu, Gangshan},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages={13608--13618},
  year={2022}
}
@misc{cui2023mixformer,
      title={MixFormer: End-to-End Tracking with Iterative Mixed Attention}, 
      author={Yutao Cui and Cheng Jiang and Gangshan Wu and Limin Wang},
      year={2023},
      eprint={2302.02814},
      archivePrefix={arXiv}
}