Skip to content

An end to end software for detecting humans in a scene, and tracking their trajectory using YOLOV3. Application developed using C++ and has been tested using unit testing.

License

Notifications You must be signed in to change notification settings

Achuthankrishna/ENPM808X-Human-Detection-and-Tracking

Repository files navigation

Human Detection and Tracking

Human Detection and Tracking Badges

CICD Workflow status codecov License

Overview

The "Human Tracker for Robots" project is a cutting-edge proposal presented to Acme Robotics aimed at addressing the challenge of "Human Detection and Tracking" within an indoor office environment using a specially designed robotic module equipped with a monocular video camera. The primary goal of this project is to develop an advanced robotic system capable of identifying and tracking humans within an office setting. This project leverages computer vision and machine learning. This is an efficient solution for various applications, such as enhancing security, improving workplace safety, and automating certain tasks within an indoor environment. Main features of the module include Human detection and Real-time Tracking. The Performance of the module is measured based on the detection accuracy and tracking precision.

Team

(Group 10)

  1. Kiran Patil
  2. Vyshnav Achutan
  3. Suryavardhan Reddy Chappidi

Phase_0 (Project Proposal)

Document Link
Project Proposal link
Quad Chart link
UML Diagram link
Activity Diagram link
Proposal Video link

Phase_1

Document Link
UML Diagram link
Proposal Video link
AIP link
Sprint Planning & Review link

Phase_2 (Implementation)

Document Link
UML Diagram link
Phase 2 Video link
AIP link
Sprint Planning & Review link

Demo

Result on Video

Result Video

Result using live camera

Result Video

Compilation

$ git clone https://github.com/Achuthankrishna/ENPM808X-Human-Detection-and-Tracking.git
$ cd ENPM808X-Human-Detection-and-Tracking
$ cmake -S ./ -B build/
$ cmake --build build/

Testing

$ cd build/
$ ctest --test-dir build/

Notes

  • The results for cpplint and cppcheck are included in the results folder.
  • Doxygen-generated docs are present in the Docs folder.

About

An end to end software for detecting humans in a scene, and tracking their trajectory using YOLOV3. Application developed using C++ and has been tested using unit testing.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published