VTSNN: A Virtual Temporal Spiking Neural Network (first image reconstruction in SNN)

Image reconstruction workflow in SNN

Encoding example

Backpropagation comparison

Reconstruction results of VTSNN under different noises

official implementation of VTSNN: A Virtual Temporal Spiking Neural Network

Accepted to Frontiers in Neuroscience (IF = 5.152)

paper: https://www.frontiersin.org/articles/10.3389/fnins.2023.1091097

Get started

git clone https://github.com/bollossom/VTSNN.git
cd VTSNN
pip install spikingjelly

Dataset and weight

Our coding MNIST dataset can be found in

link：https://pan.baidu.com/s/11DaQt1Pb9E-V60bA_vHD6Q 
code：sm15

Our MNIST weight dataset can be found in

link：https://pan.baidu.com/s/1PpuRKxmJ4mJnvmBHbSDL3w 
code：c50

If you find VTSNN module useful in your work, please cite the following source:

@article{qiu2023vtsnn,
  title={VTSNN: a virtual temporal spiking neural network},
  author={Qiu, Xue-Rui and Wang, Zhao-Rui and Luan, Zheng and Zhu, Rui-Jie and Wu, Xiao and Zhang, Ma-Lu and Deng, Liang-Jian},
  journal={Frontiers in Neuroscience},
  volume={17},
  pages={1091097},
  year={2023},
  publisher={Frontiers Media SA}
}