Welcome to my GitHub profile! I'm a passionate developer, problem-solver, and tech enthusiast. Here's a bit about me:
- 🎓 Aspiring Data Scientist, currently applying for a Summer Internship at Corsair.
- 💡 I love working on projects related to Machine Learning, Computer Vision, and Robotics.
- 🌱 Currently learning and exploring Generative Adversarial Networks (GANs), ROS2 Navigation, Go (Golang), and Rust.
- 🛠️ Strong experience in:
- Python, OpenCV, PyTorch, YOLO, ROS2
- C++ for high-performance computing
- Web development with Next.js & Material-UI
🛠️ PipeX
Pipe Classification System: Classifying different types of construction pipes using a custom YOLOv5 model.
- Tech Stack: Python, OpenCV, YOLOv5
- Highlights: Achieved high accuracy in real-world scenarios.
🐍 RL Snake
Reinforcement Learning Snake Game: An RL agent controls the snake to collect food and survive.
- Tech Stack: Python, PyTorch, NumPy, Pygame
- Highlights: Trained using Deep Q-Learning for optimal performance.
🚗 ParkEase
Parking Space Detection: Detecting available parking spaces using CCTV input.
- Tech Stack: Python, OpenCV, NumPy, MediaPipe
- Highlights: Real-time detection with high precision.
Basketball Trajectory Prediction: Predicting the trajectory of a ball using contour detection.
- Tech Stack: OpenCV, MediaPipe, NumPy
- Highlights: Enhanced accuracy for trajectory predictions.
ROS 2-Based Perception Pipeline: Contributions to an open-source project for robotics perception.
- Tech Stack: ROS 2, YOLOv8, OpenCV
- Highlights: Implemented segmentation modules for advanced robotic perception.
Drone-Based Recognition: Recognizing humans from aerial footage for tracking and analysis.
- Tech Stack: Python, OpenCV, Deep Learning
- Highlights: Optimized for drone camera feeds.
Image Sorting by Bounding Box Area: Sorting images based on the bounding box areas formed by labels.
- Tech Stack: Python, NumPy
- Highlights: Simplified dataset organization for ML tasks.
Thank you for stopping by! Feel free to explore my repositories and reach out if you'd like to collaborate. 🚀