Neuromorphic Event-based Space-Time Template Action Recognition

Neuromorphic vision hardware enables observed actions to be reduced to a series of spike trains; these trains contain unique properties relevant to observed actions. This paper presents an approach to event-based image processing which allows for the detection of specific fine grain actions through the adoption of template matching alongside neuromorphic hardware. The proposed approach was applied to the detection of breathing actions in an ambient assisted living (AAL) environment, this involved the detection of shallow, normal and heavy breathing for multiple participants using a single template. The results gained suggest that this approach could be useful when deployed in an AAL environment.

[1]  T. Delbruck,et al.  > Replace This Line with Your Paper Identification Number (double-click Here to Edit) < 1 , 2022 .

[2]  Understanding Video Latency What is video latency and why do we care about it ? , 2015 .

[3]  Arnaud Delorme,et al.  Spike-based strategies for rapid processing , 2001, Neural Networks.

[4]  George L. Malcolm,et al.  Eye Movements and Visual Encoding During Scene Perception , 2009, Psychological science.

[5]  Tobi Delbrück,et al.  Steering a predator robot using a mixed frame/event-driven convolutional neural network , 2016, 2016 Second International Conference on Event-based Control, Communication, and Signal Processing (EBCCSP).

[6]  Eli Shechtman,et al.  Space-time behavior based correlation , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).