[ICLR2025] Learning Gain Map for Inverse Tone Mapping

Yinuo Liao, Yuanshen Guan, Ruikang Xu, Jiacheng Li, Shida Sun, Zhiwei Xiong*

[Paper Link]　 [Datasets]　 [Codes]　 [Scripts]　 [Contact]

@inproceedings{Liao_2025_ICLR,
    title     = {Learning Gain Map for Inverse Tone Mapping},
    author    = {Yinuo Liao and Yuanshen Guan and Ruikang Xu and Jiacheng Li and Shida Sun and Zhiwei Xiong},
    booktitle = {The Thirteenth International Conference on Learning Representations},
    month     = {April},
    year      = {2025}
}

1. Datasets

We provide Synthetic Dataset and Real-world Dataset, which are organized by following four parts:

• image: The input SDR Images
• gainmap: The Gourd-Turth Gain Maps
• metadata: The metadata for restore HDR form SDR-GM pair (Only Qmax here)
• thumbnail: The down-sampled SDR Images in resolution 256×256 (Bicubic interpolation)

The data structure in dataset will be like:

synthetic_dataset
├── train
|   ├── image
|   |   └── *.png
|   ├── gainmap
|   |   └── *.png
|   ├── metadata
|   |   └── *.npy
|   └── thumbnail
|       └── *.png
└── test
    ├── image
    |   └── *.png
    ├── gainmap
    |   └── *.png
    ├── metadata
    |   └── *.npy
    └── thumbnail
        └── *.png

and more information can be found in the paper and the table below:

	Synthetic Dataset	Real-world Dataset
Source	HDR video frames	taken photos
Volume	ㅤㅤㅤ900 trainset & 100 testsetㅤㅤㅤ	ㅤㅤㅤ900 trainset & 100 testsetㅤㅤㅤ
SDR White Level	100 nit	203 nit
HDR Peak Level	800 nit	1015 nit
ㅤㅤQmax Rangeㅤㅤ	[0, 3] ([0, log8])	[0, 2.32] ([0,log5])
Input SDR Image	3840×2160 8bit RGB	4096×3072 8bit RGB
Gourd-turth Gain Map	3840×2160 8bit Gray	2048×1536 8bit Gray
ㅤㅤDownload Linkㅤㅤ	[BaiduNetDisk] [OneDrive] [MEGA]	[BaiduNetDisk] [OneDrive] [MEGA]

2. Codes

2.1 How to test

Please download our dataset first, then modify the dataroot in ./codes/options/config/test_syn.yml to the path you store dataset, and you can modify pretrain_model_G to choose the pretrained model. When the configuration in test_syn.yml is ready, you can run the conmand:

cd codes
python test.py -opt options/config/test_syn.yml

The test results will be saved to ./results/test_name.

2.2 How to train

To facilitate the training process, please modify the data path in crop_training_patch.py in [Scripts] and run it to crop the image and gainmap to patch:

cd scripts
python crop_training_patch.py --input_folder ../Synthetic_dataset/train/image   --save_folder ../Synthetic_dataset/train/image_sp   --n_thread 20 --crop_sz 480 --step 480
python crop_training_patch.py --input_folder ../Synthetic_dataset/train/gainmap --save_folder ../Synthetic_dataset/train/gainmap_sp --n_thread 20 --crop_sz 480 --step 480

It will generate pathes of image to image_sp folder, and the pathes of gainmap to gainmap_sp folder. After that, please modify the dataroot in ./codes/options/config/train_syn.yml to the sub-folder, then tun:

cd codes
python train.py -opt options/config/train_syn.yml

The checkpoints and training states can be found ./experiments/train_name.

3. Scripts

We provide several practical scripts in ./scripts and the details are as following:

• crop_training_patch.py: This script crops the images to patches for training. (from HDRTVNet)
• extract_double_layer_hdr.py: This script extracts sdr, gainmap and qmax from double-layer HDR file.
• render_sdr_gm_to_linear_hdr.py: This script restores linear HDR from sdr, gainmap and qmax.
• pq_visualize.py: This script converts the linear HDR to PQ-OETF encoded HDR image for visualization.

4. Contact

If you have any questions, please describe them in issues or contact [email protected]

5. Acknowledgment

We appreciate the following github repositories for their valuable work:

• BasicSR: https://github.com/xinntao/BasicSR
• HDRSample: https://github.com/JonaNorman/HDRSample
• HDR Toys: https://github.com/natural-harmonia-gropius/hdr-toys