SpikeSEG: Spiking segmentation via STDP saliency mapping

Research output: Contribution to conferencePaper

20 Downloads (Pure)

Abstract

Taking inspiration from the structure and behaviourof the human visual system and using the Transposed Convo-lution and Saliency Mapping methods of Convolutional NeuralNetworks (CNN), a spiking event-based image segmentationalgorithm, SpikeSEG is proposed. The approach makes use ofboth spike-based imaging and spike-based processing, where theimages are either standard images converted to spiking images orthey are generated directly from a neuromorphic event drivensensor, and then processed using a spiking fully convolutionalneural network. The spiking segmentation method uses the spikeactivations through time within the network to trace back anyoutputs from saliency maps, to the exact pixel location. Thisnot only gives exact pixel locations for spiking segmentation,but with low latency and computational overhead. SpikeSEGis the first spiking event-based segmentation network and overthree experiment test achieves promising results with 96%accuracy overall and a 74% mean intersection over union forthe segmentation, all within an event by event-based framework.
Original languageEnglish
Number of pages8
Publication statusPublished - 19 Jul 2020
EventInternational Joint Conference on Neural Networks: World Congress on Computational Intelligence - SEC, Glasgow, United Kingdom
Duration: 19 Jul 202024 Jul 2020
Conference number: 48605X
https://wcci2020.org/

Conference

ConferenceInternational Joint Conference on Neural Networks
Abbreviated titleIJCNN
CountryUnited Kingdom
CityGlasgow
Period19/07/2024/07/20
Internet address

Keywords

  • spiking nural network
  • SNN
  • convolution
  • segmentation

Fingerprint Dive into the research topics of 'SpikeSEG: Spiking segmentation via STDP saliency mapping'. Together they form a unique fingerprint.

  • Cite this

    Kirkland, P., Di Caterina, G., Soraghan, J., & Matich, G. (2020). SpikeSEG: Spiking segmentation via STDP saliency mapping. Paper presented at International Joint Conference on Neural Networks, Glasgow, United Kingdom.