Skip to content
No description, website, or topics provided.
Branch: master
Clone or download
Pull request Compare This branch is 1 commit ahead, 5 commits behind andresvasquezv:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
config
launch
msg
scripts
src
trained_models
CMakeLists.txt
README.md
package.xml

README.md

Hospital people detector

This ROS package provides a pipeline to detect and classify people in hospital environments.

Prerequisites

Fast R-CNN (see: Fast R-CNN installation instructions)

Running the test

First download the test bagfile wich contains a sequence collected with a Kinect V2 and Xtion from here:

wget https://drive.google.com/open?id=1mNeRLmOISqTkXUaf4OdVDLOUaAtb_5On

To run the test for Kinect camera:

roslaunch hospital_people_detector hospital_Kinect_test.launch

To run the test for Xtion:

roslaunch hospital_people_detector hospital_Xtion_test.launch

You can choose to use the RGB or DepthJet detector by modifying the parameter in the launch file:

<param name="classifier_type" value="RGB" />

or

<param name="classifier_type" value="DepthJet" />

IMPORTANT: Before you run the test you have to change the paths to Caffe and Fast R-CNN directories in the launch file:

<param name="caffe_directory" value="/home/my_user/fast-rcnn/caffe-fast-rcnn" />
<param name="fast_rcnn_directory" value="/home/my_user/fast-rcnn" />

Play the bagfile by typing:

rosbag play Downloads/test_hospital.bag --clock
You can’t perform that action at this time.