Code and dataset repository for our paper entilted "Alignment-Free RGB-T Salient Object Detection: A Large-scale Dataset and Progressive Correlation Network" accepted at AAAI 2025.
arXiv version: https://arxiv.org/pdf/2412.14576.
The model and results are available now. [17th, Jul, 2025]
Thank you for your attention.
The compressed UVT20K dataset containing the annotations of saliency maps, edges, scribbles, and challenge attributes can be found here. [baidu pan fetch code: v2rc] or [google drive]
The predicted results of our models can be found here. [baidu pan fetch code: nhau]
The parameters of our models can be found here. [baidu pan fetch code: bz6x]
The predicted results of the comparison methods can be found here. [baidu pan fetch code: 3kru]
- Download the UVT20K dataset for training and testing.
- Download the pretrained parameters of the backbone from here. [baidu pan fetch code: mad3]
- Download the pretrained parameters of the IHN model from here.
- Organize dataset and pretrained model directories.
- Create directories for the experiment and parameter files.
- Please use
condato installtorch(1.12.0) andtorchvision(0.13.0). - Install other packages:
pip install -r requirements.txt. - Set your path of all datasets in
./options.py.
python -m torch.distributed.launch --nproc_per_node=2 --master_port=2212 train_parallel.py
python test_produce_maps.py
If you think our work is helpful, please cite:
@inproceedings{wang2025alignment,
title={Alignment-Free RGB-T Salient Object Detection: A Large-scale Dataset and Progressive Correlation Network},
author={Wang, Kunpeng and Chen, Keke and Li, Chenglong and Tu, Zhengzheng and Luo, Bin},
booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
volume={39},
number={7},
pages={7780--7788},
year={2025}
}
The implement of this project is based on the following link.
If you have any questions, please contact us ([email protected]).