SSVEP_Stimuli

Code to accompany our International Conference on Robotics and Automation (ICRA) paper entitled - Using variable natural environment brain-computer interface stimuli for real-time humanoid robot navigation.

The original code for SSD is from https://github.com/amdegroot/ssd.pytorch.

The code is structured as follows:

• onebox_stimuli.py contains psychopy code for a simple SSVEP stimuli ;
• onebox_stimuli_RDA.py contains code for integration between simple SSVEP stimuli with RDA server ;
• multiclass_stimuli.py contains code for multiclass SSVEP stimuli -- 4 frequency classes example ;
• ssd_stimuli.py contains code for displaying SSD object detection in psychopy as SSVEP stimuli;
• ssd.py contains SSD architecture (original work from https://github.com/amdegroot/ssd.pytorch. ;
• ssd_predict.py contains prediction test on the image frame modified from .https://github.com/amdegroot/ssd.pytorch/blob/master/demo/live.py. ;
• ssd_stimuli_realtime.py our Varible SSVEP Stimuli code using SSD object detection in real time (on-board camera of NAO) ;

Dependencies and Requirements

The code has been designed to support python 3.6+ only. The project has the following dependencies and version requirements:

• torch=1.6.0+
• numpy=1.16++
• python=3.6.5+
• scipy=1.1.0+
• PsychoPy=3.0
• pynaoqi=2.7+
• rdaclient (https://github.com/belevtsoff/rdaclient.py)

Training weights

Please refer to https://github.com/amdegroot/ssd.pytorch on how to download the training weights.

Cite

Please cite the associated papers for this work if you use this code:

@inproceedings{aznan2019using,
  title={Using variable natural environment brain-computer interface stimuli for real-time humanoid robot navigation},
  author={Aznan, Nik Khadijah Nik and Connolly, Jason D and Al Moubayed, Noura and Breckon, Toby P},
  booktitle={2019 International Conference on Robotics and Automation (ICRA)},
  pages={4889--4895},
  year={2019},
  organization={IEEE}
}