A naive ROS interface for visualDet3D.

YOLO3D ROS Node

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

All parameters are exposed in the launch file.

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.