Home

Awesome

Chain Normalization

Efficient Spatial-Temporal Normalization in Surface of Active Events

Prerequisite

  1. Install the following packages
sudo apt-get install cmake gcc g++ libeigen3-dev
  1. Install the OpenCV

How to build?

git clone https://github.com/eleboss/chain.git
cd ./chain
mkdir build
cd ./build
cmake ../
make

How to use?

  1. Download the txt files of event data from DAVIS240C Dataset.
  2. Unzip it.
  3. Find the file event.txt, and change the file path in c++ code to it.
  4. Repeat the build process.
  5. Use the following to run
./csae
./stackchain

BTW, you can use the python scripts for visualization.

Prerequisite

pip install mayavi

Run the normalization code
Then

cd ./scripts
python mayavi_matc.py
<!-- #### Results --> <!-- ![demo](./images/demo.png) --> <!-- #### Animation --> 
@ARTICLE{9095269,
  author={S. {Lin} and F. {Xu} and X. {Wang} and W. {Yang} and L. {Yu}},
  journal={IEEE Robotics and Automation Letters}, 
  title={Efficient Spatial-Temporal Normalization of SAE Representation for Event Camera}, 
  year={2020},
  volume={5},
  number={3},
  pages={4265-4272},
  doi={10.1109/LRA.2020.2995332}}