Skip to content

Owen-Liuyuxuan/visualDet3D_ros

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

YOLO3D ROS Node

This repo contains a Monocular 3D detection Ros node. Base on https://github.com/Owen-Liuyuxuan/visualDet3D.

Update (2023-10-2): This repo now also supports inference from FSNet/VisionFactory. Use the mono3d_node.py and the mono3d.launch instead. Existing models from visualDet3D are still well-supported by yolo3d_node.py/yolo3d.launch

We have also provided a cookbook on how to utilize the open-source tool chains.

All parameters are exposed in the launch file.

image

Notice: This node only takes image and camera_info as input, the outputs are bounding boxes.

Setup

Install ROS, tested on Ubuntu 18.04, ROS melodic.

Enable rospy in Python3 (this should not affect Python2), but it does not enable tf in Python3.

sudo apt-get install python3-catkin-pkg-modules
sudo apt-get install python3-rospkg-modules

Clone this repo into a ROS workspace and run

catkin_make
source devel/setup.bash

under the workspace folder.

Also modify the launch file.

Subscribed Topics

image_raw (sensor_msgs/Image)

A stream of rectifiled image to be predicted using monodepth.

camera_info (sensor_msgs/CameraInfo)

Camera calibration information of the rectified image.

Published Topics

bboxes (visualization_msgs/MarkerArray)

Predicted bboxes 3D objects.

About

A naive ROS interface for visualDet3D.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published