human-detection-and-tracking-cpp

This project is designed and programmed to detect and track humans which will provide the location of the humans in a robot's frame of reference. Agile Iterative Process is used for the development of this project consisting of two sprints.

Authors

Sprint 1 -

• Madhu Narra Chittibabu (Driver) (UID: 118206196)
• Sharmitha Ganesan (Navigator) (UID: 117518931)
• Koundinya Vinnakota (Design Keeper) (UID: 118239811)

Sprint 2 -

• Madhu Narra Chittibabu (Navigator) (UID: 118206196)
• Sharmitha Ganesan (Driver) (UID: 117518931)
• Koundinya Vinnakota (Design Keeper) (UID: 118239811)

Overview

Human Detection is a branch of Object Detection in Computer Vision. Object Detection is the task of identifying the presence of predefined types of objects in an image. This task involves both identification of the presence of the objects and identification of the rectangular boundary surrounding each object (i.e. Object Localisation). An object detection system which can detect the class “Human” can work as a Human Detection System.

Purpose

We aim to design and deliver a robust robust human detector and tracker using a monocular camera, directly usable in a robot’s reference frame according to the requirement specifications provided to us by ACME robotics's RnD division for integration into a future product.

Our system is built using C++ and will employ the robust YOLOv5 neural network model trained on the COCO dataset for human detection and tracking as it is one of the most accurate real-time object detection algorithms. An image from a monocular camera is pre-processed and passed to the model which outputs the location info in the image frame. It is then converted to the camera frame by using the calibration constants and then transformed into the robot's frame.

Output

Requirements

• C++ 14++
• Cmake min 3.2.1
• Opencv For installation

REFER TO THE INSTALL_DEPENDENCIES.SH FILE WHICH HAS THE COMMANDS REQUIRED FOR OUR API TO RUN.

Continuous integration and code coverage is tracked by using github workflows.

Build and Running Instructions

    git clone https://github.com/Madhunc5229/human-detection-tracking-cpp.git
    mkdir build && cd build
    cmake .. 
    make

    <!-- Run app -->
    ./app/myApp

    <!-- Run tests -->
    ./test/cpp-test

Building for code coverage

sudo apt-get install lcov
cmake -D COVERAGE=ON -D CMAKE_BUILD_TYPE=Debug ../
make
make code_coverage

This generates a index.html page in the build/coverage sub-directory that can be viewed locally in a web browser.

Run and save cpplint and cppcheck

cpplint $( find . -name *.cpp | grep -vE -e "^./build/" -e "^./vendor/") $( find . -name *.h | grep -vE -e "^./build/" -e "^./vendor/") > results/cpplint.txt

cppcheck --enable=all --language=c++ --std=c++11 -I include/ --suppress=missingIncludeSystem $( find . -name *.cpp | grep -vE -e "^./build/" -e "^./vendor/" ) $( find . -name *.h | grep -vE -e "^./build/" -e "^./vendor/") > results/cppcheck.txt

The Quad Chart for the project can be found here

The Proposal for the project can be found here

The Backlog tables for the project can be found here

The video explanation of the phase zero of this project can be found here

The video explanation of the phase one of this project can be found here

The video explanation of the phase two of this project can be found here