Deep Tensor Admm-Net for Video SCI

This repository is the implementation of Deep Tensor ADMM-Net for Snapshot Compressive Imaging.

The last test is in the environment with following additional packages:

1. Tensorflow-gpu==1.13.1 (conda install tensorflow-gpu=1.13.1)
2. yaml (pip install pyyaml)
3. hdf5storage (pip install hdf5storage)

For fast implementation, please run test.py to generate the recovered frames, stored in Result/Validation-Result/NBA_Frame.

We will integrate the model as well as data of other scenarios/dataset into this version asap. Thank you for your patience.