Neural Coding Strategies for Event-Based Vision Data

Shane Harrigan, Sonya Coleman, Dermot Kerr, Pratheepan Yogarajah, Zheng Fang, Chengdong WuICASSP 2020 · OpenReview

Abstract

Neural coding schemes are powerful tools used within neuroscience. This paper introduces three different neural coding scheme formations for event-based vision data which are designed to emulate the neural behaviour exhibited by neurons under stimuli. Presented are phase-of-firing and two sparse neural coding schemes. It is determined that machine learning approaches, i.e. Convolutional Neural Network combined with a Stacked Autoencoder network, produce powerful descriptors of the patterns within events. These coding schemes are deployed in an existing action recognition template and evaluated using two popular event-based data sets.
BibTeX
@inproceedings{harrigan2020neural,
  author = {Shane Harrigan and Sonya Coleman and Dermot Kerr and Pratheepan Yogarajah and Zheng Fang and Chengdong Wu},
  title = {{Neural Coding Strategies for Event-Based Vision Data}},
  booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
  year = {2020},
  url = {https://openreview.net/forum?id=Tv0r__haWx},
}

Loading related content...

Figure