Skip to content

Latest commit

 

History

History
51 lines (38 loc) · 1.55 KB

README.md

File metadata and controls

51 lines (38 loc) · 1.55 KB

Position Estimation of recognized objects using YOLOv7 and the Stereo Camera Model.

In this project we explain the development, implementation and integration of a Position Estimation system of recognized objects using YOLOv7. This project is divided into 2 parts, Object Recognition using the artificial vision algorithm YOLOv7 and Pose Estimation based on a Pinhole Camera Model using a ZED 2i camera.

Authors

  • Jorge Askur Vazquez Fernandez
  • Jose Miguel Flores Gonzalez

Dependencies

The rest of the required libraries are:

  pip install statistics
  pip install opencv-python
  pip install opencv-contrib-python
  sudo apt-get install ros-noetic-image-geometry
  sudo apt-get install ros-noetic-sensor-msgs
  sudo apt-get install ros-noetic-cv-bridge
  sudo apt-get install ros-noetic-visualization-msgs
  sudo apt-get install python3-catkin-tools

Install

  git clone https://github.com/JorgeAskur/PinholeCamera.git
  catkin build

Usage

#Launch ZED wrapper
roslaunch zed_wrapper zed2i.launch

#In another terminal
#Launch YOLO and Position Estimation
roslaunch yolov7_ros yolov7.launch

Additional Documentation

Link to technical report

Link to video