Projects

Here are some PERSONAL projects of mine that I have completed. As of now, I have added 5 projects in the current site. Please read the blog-posts to learn more, and feel free to contact me if you have any questions or want to discuss anything.

This Site + My Blog

HTML, Javscript, CSS, React, MongoDB, Express.js, Node.js, Bootstrap, and Gatsby

This site is a project of mine using HTML, JavaScript, and CSS along with Gatsby, React, and Bootstrap. A version of this site using the MERN (MongoDB, Express.js, React, and Node.js) stack can be found on my Github. Some older versions that were less responsive can also be found on my blog and Github. I've also completed other web-dev projects including this website for the Charette USNSCC Division. Read more about the personal touches in this blog!
TLDR: various web-dev projects you can find on my Github

Github

📝 Blog Post

Mathematics of Computer Vision

Python, OpenCV, Numpy, Pandas, Latex, Matlab

This project is an educational resource for learning Computer Vision consisting of 60+ pages of TexEd up notes and 10 Python Demos (Documented Jupyter Notebooks). These notes were presented at various FTC robotics conferences. Some relevant topics covered include the PnP Model, object detection, facial recognition, digital signals processing, image compression, and CNNs/Deep-Learning. Please read the README on the Github Repository before the notes!
TLDR: textbook-style notes written in Latex about CV + demos

Github

📝 Blog Post

✍️ Access Notes

Sparsity

C, C++, Python

This project contains implementations of Compressed Sparse Row Matrix representation (CSR), various matrix methods (multiplication, QR/LU/Cholesky Decomposition, etc), a Collaborative-Filtering scheme using my implementation of CSR (prediction scheme for content used by platforms like Netflix). It's implemented in C and has comparable run-times to similar functionalities of scipy.sparse, Python's standard module for sparse matrices. It is inspired by Dr. Bahar Asgari's PhD Thesis.
TLDR: library written in C wrapped with Python for sparse data

Github

📝 Blog Post

Surveillance Detector

Python, YOLO, PyTorch, TKinter

Human Detector uses your webcam to monitor for unauthorized access of your computer. When a person is detected, it prompts for a password and, if not entered, triggers an alarm, takes a photo, and can initiate other configurable actions like shutting down the computer. It uses a trained-version of the YOLO Object Detection System and contains thoroughly documented code on use/design of the YOLO Architecture!
TLDR: human security detector with alarm, password, and other features

Github

📝 Blog Post

Vehicle Type

Python, YOLO, PyTorch, Scikit-Learn, Pandas, Numpy

This project is a vehicle-classifier which classifies vehicles into multiple different categories based on images. It will be used in the future to analyze videos of vehicles and assess their speed and construct a "safety-score" on how the drivers are supposedly driving. It is intended to be a deep-learning based tools for traffic-cameras to learn and comment more about vehicles.
TLDR: vehicle detector for analyzing cars/vechiles and traffic cameras

Github

📝 Blog Post

CMSC320 Final Project

Jupyter Notebook, Numpy, Scikit-Learn, Hypothesis-Testing, Data Exploration, Statistics

This project is a notebook which you can find some answers to questions regarding Capital Bikeshare usage in the DMV (District of Columbia, Maryland, Virginia) area. The project goes through various data preprocesing techniques, data exploration techniques, hypothesis testing techniques, machine learning techniques, and data insights in order to further increase bike ridership written in a teaching style. This was our final group project for CMSC320 course -- Introduction to Data Science.

Github

📝 Blog Post

✍️ Access IPYNB

Hello, World!

Hi, I'm Abhi!

This Site + My Blog

Mathematics of Computer Vision

Sparsity

Surveillance Detector

Vehicle Type

CMSC320 Final Project

See ya!