Attention-based Fusion Router for RGBT Tracking. arXiv
We use the LasHeR training set for training, GTOT, RGBT210, RGBT234, LasHeR testing set, VTUAVST for testing, and their project addresses are as follows:
Clone repo:
git clone https://github.com/Alexadlu/AFter.git
cd AFter
Our code is trained and tested with Python == 3.8, PyTorch == 1.8.1 and CUDA == 11.2 on NVIDIA GeForce RTX 4090, you may use a different version according to your GPU.
conda create -n after python=3.8.13
conda activate after
pip install -r requirements.txt
- Modify the project path and dataset path in
$PROJECT_ROOT$/ltr/admin/local.py. - Download ToMP-50 pretrained weights and put it under
$PROJECT_ROOT$/ltr/models/pretrained. - Run the following command.
python ltr/run_training.py --train_module tomp --train_name tomp50_v1
- Modify the dataset path in
$PROJECT_ROOT$/pytracking/evaluation/environment.py - Put the checkpoint into
$PROJECT_ROOT$/pytracking/networksand select the checkpoint name in$PROJECT_ROOT$/pytracking/parameter/tomp/tomp50.py. Or just modify the checkpoint path in$PROJECT_ROOT$/pytracking/parameter/tomp/tomp50.py. - Run the following command.
python pytracking/run_tracker.py --tracker_name tomp --tracker_param tomp50 --runid 8600 --dataset_name lashertestingset
| Model | GTOT(PR/SR) | RGBT210(PR/SR) | RGBT234(PR/SR) | LasHeR(PR/NPR/SR) | VTUAV(PR/SR) | Checkpoint | Raw Result |
|---|---|---|---|---|---|---|---|
| AFter | 91.6 / 78.5 | 87.6 / 63.5 | 90.1 / 66.7 | 70.3 / 65.8 / 55.1 | 84.9 / 72.5 | download | download |
Please kindly cite this paper in your publications if it helps your research:
@article{Lu2025AFTER,
title={AFTER: Attention-based Fusion Router for RGBT Tracking},
author={Andong Lu and Wanyu Wang and Chenglong Li and Jin Tang and Bin Luo},
journal={IEEE Transactions on Image Processing},
year={2025},
}
Contact: adlu_ah@foxmail.com
Our project is based on the pytracking framework and ToMP. Thanks for their contributions which help us to quickly implement our ideas.