Lightweight and intuitive package designed to streamline the integration process of dotta biometric service API

comparison of different classification techniques for gender based on facial images. Using the best model for live gender classification from webcam

My Programming Assignments from the MA 544 Numerical Linear Algebra for Big Data Course from Stevens Institute of Technology

Visual Identity TUNnel. Encrypted TUN Communication with Facial-AES Key Auth.

Demo Day - Web Interface for Senior Design (2023), Texas A&M University. Team: Roshan Thomas, Fathima Hakeem, Farah Shabbir, Moussa Judia

An age estimation system that uses both image and audio analysis to predict age groups. The system combines computer vision and audio processing techniques to provide accurate age predictions through multiple input modalities.

This program demonstrates how to plot face feature vectors on a 2-D chart using t-SNE method to compress dimension of the vectors

cascade correlation filter for super fast face landmark detection (500 landmarks/sec)

A collection of my completed projects, presented in a user-friendly tabular format within the README file. Each project is accompanied by a concise description, a list of the tools and languages/libraries utilized, and a corresponding repository link for seamless navigation.

Facial feature detection using AdaBoost with shape constraints

The idea here was to create a centralised database using machine learning over all existing databases of a person's information, based on his face picture as ID, to remove the need for any physical ID proofs - your face is your unique ID proof.

An example implementation of a real-time facial recognition server with OpenVidu, Node.js and Python.

shortfuse on CelebA dataset

NTU Hackathon 2nd Prize

Hierarchical CNN based Random Forests (HCRF) approach for face super-resolution.

A curated list of weakly supervised learning methods and datasets for facial behavior analysis.

The project, funded by the National Undergrad Innovation Training Program, aims to research non-contact heart rate measurement within a year. It utilizes device cameras (e.g., computers, smartphones) to extract facial video light signals, analyze them, and determine real-time heart rates based on Imaging Photoplethysmography (IPPG).

Facial keypoints detection using CNN