Linux | |
Windows |
Find-Object project, visit the home page for more information.
Binaries:
sudo apt-get install ros-$ROS_DISTRO-find-object-2d
Source:
- To include
xfeatures2d
and/ornonfree
modules of OpenCV, to avoid conflicts withcv_bridge
, build same OpenCV version that is used bycv_bridge
. Install it in/usr/local
(default).
cd ~/catkin_ws
git clone https://github.com/introlab/find-object.git src/find_object_2d
catkin_make
roscore
# Launch your preferred usb camera driver
rosrun uvc_camera uvc_camera_node
rosrun find_object_2d find_object_2d image:=image_raw
See find_object_2d for more information.
Binaries:
To come...
Source:
cd ~/ros2_ws
git clone https://github.com/introlab/find-object.git src/find_object_2d
colcon build
# Launch your preferred usb camera driver
ros2 launch realsense2_camera rs_launch.py
# Launch find_object_2d node:
ros2 launch find_object_2d find_object_2d.launch.py image:=/camera/color/image_raw
# Draw objects detected on an image:
ros2 run find_object_2d print_objects_detected --ros-args -r image:=/camera/color/image_raw
A RGB-D camera is required. Example with Realsense D400 camera:
# Launch your preferred usb camera driver
ros2 launch realsense2_camera rs_launch.py align_depth.enable:=true
# Launch find_object_2d node:
ros2 launch find_object_2d find_object_3d.launch.py \
rgb_topic:=/camera/color/image_raw \
depth_topic:=/camera/aligned_depth_to_color/image_raw \
camera_info_topic:=/camera/color/camera_info
# Show 3D pose in camera frame:
ros2 run find_object_2d tf_example
See find_object_2d for more information (same parameters/topics are used between ROS1 and ROS2 versions).