DenseFusion_ROS

This repository is based on https://github.com/j96w/DenseFusion and https://github.com/ooooverflow/BiSeNet.

If you are a person using Docker, https://hub.docker.com/repository/docker/choo2969/ros-densefusion

segmentation weight file link densefusion weight file link

$ docker pull choo2969/ros-densefusion

Requirements

• ROS (Kinetic)
• Python2.7
• Pytorch 0.4.1
• PIL
• scipy
• numpy
• pyyaml
• logging
• matplotlib
• CUDA

Start

we have tested on Ubuntu 16.04 with ROS Kinetic and NVIDIA Titan XP and Geforce 1080 Ti

1. Start camera node (D435)

   • Step1. Run your own camera, If your camera is not a D435 or D415, you will need to edit the RGB image and Depth Subscriber. Edit image_subscriber and depth_subscriber with your camera node

     vim path/densefusion/scripts/experiments/scripts/ros_eval_msg.sh

   • Step2. Edit the cam_cx,cam_cy,cam_fx,cam_fy values

     vim path/densefusion/scripts/tool/ros_eval_ycb_message.py

2. Start

   sh path/densefusion/scripts/experiments/scripts/ros_eval_msg.sh

   Running this whill launch the SErvice Sever rining 6D Pose Estimation