addressalign-toparrow-leftarrow-rightbackbellblockcalendarcameraccwcheckchevron-downchevron-leftchevron-rightchevron-small-downchevron-small-leftchevron-small-rightchevron-small-upchevron-upcircle-with-checkcircle-with-crosscircle-with-pluscrossdots-three-verticaleditemptyheartexporteye-with-lineeyefacebookfolderfullheartglobegmailgooglegroupsimageimagesinstagramlinklocation-pinm-swarmSearchmailmessagesminusmoremuplabelShape 3 + Rectangle 1outlookpersonJoin Group on CardStartprice-ribbonImported LayersImported LayersImported Layersshieldstartickettrashtriangle-downtriangle-uptwitteruseryahoo

Object Tracking Challenge 3

We'll continue our work on object tracking.  The goal remains to write code to identify a foreground object against a background, from a moving camera (e.g. on a robot).

See the Object Tracking Challenge blog post for more details:

As usual, we'll spend the first hour on presentations of demos and code people have written for the challenge (or just cool computer vision hacks). The second hour will be informal discussion and helping people with OpenCV problems.

Join or login to comment.

  • JJ S.

    Thanks to John for hosting this event.

    John demoed his progress on the challenge with a loop for finding the Harris Corners and then using the Lucas-Kanade method to distiquish between the forground object and the background scene.

    Koshy showed his method of using HOG to disguish cars passing by, across the viewport. He was able to show the cars highlighted in yellow but the artifacts were in red and green.

    Albert demoed his Metagoo Product: Saybooks. This provides object identification and annotation as well as Augmented Reality tinkering.

    Again, thank you so much and nice to meet you all.


    June 3, 2014

    • koshy g.

      thanks for the report

      June 3, 2014

  • JJ S.

    1000 char limit? What is this are we in the 80's?

    John also demoed his ResisterVision iphone app which identifies the resistance value by analyzing the color bands. He noted that distinguishing the similar colors, especially with iphone5's changing torch hues, is a bit difficult. He realized that there are some combinations of colors which do not exist and this does help the prediction.

    John also demoed a Galileo/iPhone app which tracks the orange object (a roadworker's safety vest which John vested) as he moved about the room.

    Finally, John put us to the challenge, due in two weeks :)


    June 3, 2014

16 went

People in this
Meetup are also in:

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy