hello-trouble / YoloV5_D415_Ros_Topic

YoloV5_D415_ROS

The code from YoloV5_D415 (https://github.com/Jumbo-zczlbj0/YoloV5_D415) has been modified. Speech recognition and multithreading features were removed, and an output for object category and depth in the ROS topic was added

Install

1.Install cuda && pytorch

install CUDA URL：https://developer.nvidia.com/cuda-toolkit

install pytorch URL：https://pytorch.org/

2.git && install requirements

mkdir -p ~/catkin_ws/src

cd ~/catkin_ws/src/

git clone https://github.com/ultralytics/yolov5.git

cd ./YoloV5_D415_Ros_Topic/scripts

chmod +x pyrealsense2_camera.py

yolov5:

git clone https://github.com/ultralytics/yolov5.git

pip3 install requirements.txt

3.Install Intel RealSense SDK 2.0

Choose the version that suits you based on your computer system and install SDK-2：https://www.intelrealsense.com/sdk-2/

Install pyrealsense2：pip install pyrealsense2

4.ROS

Ubuntu20.04 + ROS: https://wiki.ros.org/noetic/Installation/Ubuntu

5.Build the ROS package:

cd ~/catkin_ws

catkin_make

Run

cd ~/catkin_ws

source ~/catkin_ws/devel/setup.bash

roslaunch yolo_topic yolo_topic.launch