Home

Awesome

Self-Holo

Diffraction model-informed neural network for unsupervised layer-based computer-generated holography.<br>

Paper

X. Shui, H. Zheng, X. Xia, F. Yang, W. Wang, and Y. Yu, “Diffraction model-informed neural network for <br> unsupervised layer-based computer-generated holography,” Opt. Express 35(25), (2022).

Dataset

The RGB-D datasets are from TensorHolography.

High-level Structure

The code is organized as follows:

./src/

We recommend that the readers to experiment with different upsampling approaches or different CNN frameworks<br> to further improve the quality of the images.

Running the test

python ./src/train.py  --run_id=selfholo

Ackonwledgement

We are thankful for the open source of NeuralHolography, HoloEncoder,and HoloEncoder-Pytorch-Version. These works are very helpful for our research.