Uncertainty Guided Refinement for Fine-grained Salient Object Detection

Source code of 'Uncertainty Guided Refinement for Fine-grained Salient Object Detection', which is accepted by TIP 2025. You can check the manuscript on Arxiv or IEEE.

Environment

Python 3.9.13 and Pytorch 1.11.0. Details can be found in requirements.txt.

Data Preparation

All datasets used can be downloaded at here [arrr].

Training set

We use the training set of DUTS to train our UGRAN.

Testing Set

We use the testing set of DUTS, ECSSD, HKU-IS, PASCAL-S, DUT-O, and SOD to test our UGRAN. After downloading, put them into /datasets folder.

Your /datasets folder should look like this:

-- datasets
   |-- DUT-O
   |   |--imgs
   |   |--gt
   |-- DUTS-TR
   |   |--imgs
   |   |--gt
   |-- ECSSD
   |   |--imgs
   |   |--gt
   ...

Training and Testing

1. Download the pretrained backbone weights and put them into pretrained_model/ folder. ResNet [uxcz], SwinTransformer are currently supported.

2. Run python train_test.py --train=True --test=True --eval=True --record='record.txt' for training and testing. The predictions will be in preds/ folder and the training records will be in record.txt file.

Evaluation

Pre-calculated saliency maps: UGRAN-R [b7fx], UGRAN-S [gfxr]
Pre-trained weights: UGRAN-R [c3eq], UGRAN-S [n7tr]

For PR curve and F curve, we use the code provided by this repo: [BASNet, CVPR-2019].
For MAE, Weighted F measure, E score, and S score, we use the code provided by this repo: [PySODMetrics].

Evaluation Results

Quantitative Evaluation

Precision-recall and F-measure curves

Visual Comparison

Acknowledgement

Our idea is inspired by InSPyReNet and MiNet. Thanks for their excellent work. We also appreciate the data loading and enhancement code provided by plemeri, as well as the efficient evaluation tool provided by lartpang.

Citation

If you think our work is helpful, please cite

@ARTICLE{10960487,
  author={Yuan, Yao and Gao, Pan and Dai, Qun and Qin, Jie and Xiang, Wei},
  journal={IEEE Transactions on Image Processing}, 
  title={Uncertainty-Guided Refinement for Fine-Grained Salient Object Detection}, 
  year={2025},
  volume={34},
  number={},
  pages={2301-2314},
  doi={10.1109/TIP.2025.3557562}}