My Projects

Technical Showdown!

A showcase of the technical projects that I am proud of.

  • German Street Signs predictor using Tensorflow 2

    In this project, I have created a deep learning model that interprets what a certain street sign means. For this purpose, the GTSRB - German Traffic Sign Recognition Benchmark dataset has been used. Along with this, the model has been implemented in three different ways as well. Last, but not the least, the "Hello, World!" of deep learning i.e., training the model with the MNIST dataset has also been performed.

  • Computer Vision operations using OpenCV

    In this project, the following operations, often used in beginning or improving Computer Vision and Image Processing related experiments are performed. The facilities provided by OpenCV are used for this purpose:

    1. Pinhole camera model and calibration
    2. Stereo calibration and rectification
    3. Sparse depth triangulation
    4. Dense depth triangulation

  • Discrete Kalman Filter for 1D and 2D object tracking

    In this project, I have used various C++ and OOP concepts to implement a Kalman Filter for tracking an object in 1-D in C++. This project gave me exposure to implementing a Kalman Filter on a custome-defined model track system in C++. The highlight of this project was using the Eigen library for linear algebraic operations, Matplot++ plotting library for plotting graphs, OpenCV4 library for Object detection that would be used in the object tracking and CMake for controlling the software compilation process.

  • Computational Geometry algorithm implementations in C++ (WIP)

    In this project, I have used various C++ and OOP concepts to implement Computational Geometry algorithms in C++. This project gave me exposure to not only coding genuine projects in C++, but also to using CMake for build system generation and using the GoogleTest framework for unit testing. The algorithms deployed in this project are the very basics of computer graphics and collision avoidance systems used for SLAM.