Attention: Answers.ros.org is deprecated as of August the 11th, 2023. Please visit robotics.stackexchange.com to ask a new question. This site will remain online in read-only mode during the transition and into the foreseeable future. Selected questions and answers have been migrated, and redirects have been put in place to direct users to the corresponding questions on Robotics Stack Exchange. Additional details are available here.

ppl_detection package and training data

asked 2012-04-09 22:03:35 -0600

Grega Pusnik

460 ●25 ●30 ●43 http://si.linkedin.com...

updated 2012-04-10 12:35:28 -0600

David Lu

10932 ●123 ●219 ●250 http://MetroRobots.com/

Do you know if you have to train ppl_detection package or should work at it's best out of the box. Because I get very poor results. Almost every object is human.

Thank you.

edit retag flag offensive close merge delete

add a comment

1 Answer

Sort by » oldest newest most voted

answered 2012-09-24 17:20:49 -0600

James Diprose
123 ●6 ●9 ●14

updated 2012-09-25 13:41:12 -0600

Its labelling everything as a person for me as well (see the image).

But I have a solution, don't use ppl_detection, use openni_tracker with the latest openni and nite libraries instead. It can automatically find people, is orders of magnitude more accurate than ppl_detection and tracks skeleton data too. This is a good tutorial on how to set it up.

image description

edit flag offensive delete link

add a comment

Question Tools

1 follower

Stats

Asked: 2012-04-09 22:03:35 -0600

Seen: 497 times

Last updated: Sep 25 '12

ppl_detection package and training data edit

1 Answer

Question Tools

Stats

Related questions

ppl_detection package and training data