Awesome
AHDRNet
<small> Attention-guided Network for Ghost-free High Dynamic Range Imaging (AHDR)
Qingsen Yan*, Dong Gong*, Qinfeng Shi, Anton van den Hengel, Chunhua Shen, Ian Reid, Yanning Zhang. In IEEE Conference on Compute rVision and Pattern Recognition (CVPR), 2019:1751-1760. (* Equall contribution) [Paper][Project]
<img src='imgs/frame.jpg' width=790>Requirements
- Python 2.7
- PyTorch 0.3.1 (tested with 0.3.1)
- MATLAB (for data preparation)
Usage
Testing
- Install this repository and the required packages. A pretrained model is in
./trained-model
. - Prepare dataset.
- Download dataset.
- Move the dataset into
./dataset
. - Processed dataset can be obtained by running the corresponding script in
./matlab_script (To be added)
.
- Run
python script_testing.py
files.
Training
Training code will be released later.
Examples of the Results
<img src='imgs/fig.jpg' width=420>Examples of the Estimated Attention Maps
<img src='imgs/att_map.jpg' width=420>Citation
If you use this code for your research, please cite our paper.
@article{yan2019attention,
title={Attention-guided Network for Ghost-free High Dynamic Range Imaging},
author={Yan, Qingsen and Gong, Dong and Shi, Qinfeng and Hengel, Anton van den and Shen, Chunhua and Reid, Ian and Zhang, Yanning},
journal={IEEE Conference on Compute rVision and Pattern Recognition (CVPR)},
year={2019}
pages={1751-1760}
}