Car Detection Using Fast Region-based Convolutional Networks (R-CNN) in ROS with Caffe
To run this, you need to install Fast-RCNN and Autoware. Just in case you got error regarding hd5f when making Fast-RCNN, add the following lines to your Makefile.config
1 2 |
INCLUDE_DIRS += /usr/include/hdf5/serial LIBRARY_DIRS += /usr/lib/x86_64-linux-gnu/hdf5/serial |
Now run the following command to start:
1 2 3 4 |
source /opt/ros/kinetic/setup.bash source /home/<user_name>/Autoware/ros/devel/setup.sh export ROS_PACKAGE_PATH=/home/<user_name>/Autoware/:$ROS_PACKAGE_PATH roslaunch cv_tracker rcnn.launch |
if you got an error like :
1 |
Check failed: error == cudaSuccess (30 vs. 0) unknown error |
That means your graphics card is not ready or accessible, in my everytime I suspend my notebook I get that …
Car Detection Using Fast Region-based Convolutional Networks (R-CNN) in ROS with Caffe Read More »