ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange
Ask Your Question

ReIn-like framework for OpenCV+ROS+Python?

asked 2011-05-13 09:26:30 -0500

Pi Robot gravatar image

updated 2011-05-13 09:48:10 -0500


I am working on using ROS + OpenCV via the cv_bridge package to process 2D video images using a sequence of filters. One example is finding the Good Features to Track (GFTT) in a face by first doing face detection, then passing the returned bounding box as the ROI for GFTT. Right now I am just combining these filters within a single node (Python script) but I was looking at the rein package/library and was wondering if something similar has or is been done for OpenCV and Python?


edit retag flag offensive close merge delete

1 Answer

Sort by ยป oldest newest most voted

answered 2011-06-17 11:38:21 -0500

ethanrublee gravatar image

Your approach is completely appropriate. rein should be considered no man's land and is unsupported at the moment.

edit flag offensive delete link more


Thanks for the response. I've got a got a good start on a more modular vision framework similar to rein and will post back once I have something to show.
Pi Robot gravatar image Pi Robot  ( 2011-06-18 01:37:51 -0500 )edit
We are working on an OpenCV + Python graph based modular vision framework, that will be released with docs in a couple of weeks. Its more low level then "object recognition", but does enable higher level vision pipelines.
ethanrublee gravatar image ethanrublee  ( 2011-06-18 05:12:59 -0500 )edit
Sounds exactly what I was looking for. I'll put my own work along those lines on hold until I see what you have developed.
Pi Robot gravatar image Pi Robot  ( 2011-06-18 11:49:32 -0500 )edit

Question Tools


Asked: 2011-05-13 09:26:30 -0500

Seen: 396 times

Last updated: Jun 17 '11