My Projects
01
Open CV object detection
I developed a real-time detection system using OpenCV for image processing, MediaPipe for hand tracking, and YOLOv5 with PyTorch for object detection. The system integrates Haar Cascade for face detection, blending classical and modern techniques, and leverages custom multi-threading in C++ and Python for high-performance applications.
02
Open CV 2D Games
This project uses computer vision and gesture recognition with OpenCV and MediaPipe to create interactive games like Rock-Paper-Scissors and Flappy Bird, controlled via real-time hand gestures. It showcases gesture-based interfaces’ potential for immersive gaming and accessibility, enabling hands-free interaction for individuals with mobility impairments.
03
Gestura - Gesture Controlled System
It is innovative application that allows users to control their computer's system volume and mouse pointer through simple hand gestures. Utilizing computer vision (OpenCV) technology and hand tracking via a webcam, the software interprets finger movements to adjust the volume or move the mouse cursor seamlessly.
04
Image Classification - Sorting Objects
This project automates tomato ripeness classification using a custom CNN and transfer learning with ResNet50, leveraging PyTorch and torchvision. The model, trained on 177 labeled images, distinguishes ripe from unripe tomatoes with high accuracy. Integrated Gradients explain feature attribution, enhancing interpretability. The system reduces manual sorting labor and boosts efficiency.
05
Chatbot (Mental Health)
This Mental Health ChatBot trained on 2k+ intents is a web-based application designed to assist users by providing supportive and empathetic responses related to mental health. The chatbot uses Natural Language Processing (NLP) techniques to understand and respond to user inputs, aiming to offer guidance, comfort, and general information about mental well-being.
06
Automated Trading System (Hardware Based)
Developed and deployed an algorithmic trading bot that runs 24/7 on a Raspberry Pi. The bot implements a simple moving average crossover strategy to trade stocks and cryptocurrencies. The project involved data fetching, strategy implementation, backtesting, and automation.
