Cognitive Neuromorphic Engineering Workshop

User Tools

Site Tools


Crash course on using DVS/DAVIS event cameras

This tutorial will show the variety of options for using event event cameras with different software packages (jAER, cAER, ROS, python) but will focus on using jAER.

Plan for tutorial:

4x45m afternoon sessions to show

  1. Using these sensors in jAER (viewing, recording, playing back)
  2. Controlling the sensor parameters (“biasing”) to set event threshold, pixel bandwidth, and refractory period
  3. Using pre-built event filters to filter out noise, look at event statistics, and track simple objects
  4. Developing new event sensor algorithms in jAER.


Get USB thumb drive from Tobi to obtain working copy of jAER git . Or else, if you are brave, clone the working copy from . Note: this is a large clone of about 500MB so it is much better for the workshop to copy a zip from another participant.

I will put this large folder of about 10GB onto the sambda network drive ASAP. It has java+netbeans installers, sample data, and other tools.

For workgroup quick start (not for development)

Install a java runtime environment (JRE) and unzip the folder. Then you should be able to run the launcher jaerviewer_win64.exe or to start the AEViewer.

Possible projects

  1. Record background noise under variety of background illumination levels and sensor parameters (particularly pixel bandwidth) to make a new sensor noise dataset
  2. Implement software noise filter based on paper from UCSD: Khodamoradi, A., and Kastner, R. (2018). O(N)-Space Spatiotemporal Filter for Reducing Noise in Neuromorphic Vision Sensors. IEEE Transactions on Emerging Topics in Computing PP, 1–1. doi:10.1109/TETC.2017.2788865,
  3. Record life of CNE hotel bar over entire day from Sala Promenade room
  4. ???

Clones of most of the below packages are included on the USB thumb drive.

Event camera R&D prototypes:

cc18/dvsdavis-event-camera-crash-course/overview.txt · Last modified: 2019/05/16 20:20 (external edit)