🚗 Detect driver distractions in images for improved road safety by predicting behavior like texting, talking, and reaching behind the wheel.

A robust and explainable deepfake detection framework that integrates CNN-based feature extraction with physiological signal analysis. This project combines ResNet50, MesoNet4 Analysis module to improve the accuracy and generalizability of deepfake forensics.

This project is a convolutional neural network built in Keras/TensorFlow and trained on Caltech-101, It is integrated with my YOLO object detection model at the website below, but can also be tested separately. This code is for view and test only, If you want to use it in real projects read the LICENSE.

Convolutional neural network (CNN) from scratch for car license plate detection.

Catch and inspect rare Pokémon card misprints using image alignment and eBay-powered crawling.

An AI powered Blogging Platform - featuring a user based recommendation engine, ensemble based text classification model, LLM based content enhancer, text to image generator and a AI powered profanity filter

This is a repo comprising of S7 DL lab Programs. Programs are designed for educational purposes and demonstrate core Neural Network and Deep Learning concepts.

This is a production-ready FastAPI microservice for lung cancer detection from medical images using AI/ML models. It's the prediction API component of a larger lung cancer detection system. The service is built with a clean architecture, is containerized with Docker, and provides endpoints for prediction, health checks, and model information.

A generalized research workbench for fusing visual (CNN) and textual (RNN/LSTM) features. It provides a clean, modular Keras framework, optimized for efficient training on Google Colab by leveraging local I/O and persistent asset saving to Google Drive.

🎭 Real-time facial emotion detection using CNN trained on FER2013 dataset. Built with TensorFlow & Streamlit. Detects 7 emotions: Angry, Disgust, Fear, Happy, Sad, Surprise, Neutral. Live demo available! 😊📊⚡

Internship project on Lung Cancer Detection focused on analyzing textual clinical data, with additional insights from medical images.

Brain tumor classification using 2 pretrained CNN models and late fusion

This project builds a Convolutional Neural Network (CNN) model to recognize handwritten digits. The model is trained on a dataset of handwritten digits (like MNIST) and can accurately classify digits from 0 to 9.

COVID-19 X-ray Prediction using CNN: A deep learning model that classifies chest X-ray images into COVID-19, Pneumonia, or Normal. Utilizes Keras for preprocessing, training, and prediction, enabling rapid and accurate diagnosis support for healthcare applications.

The model is built using CNN model, the model classify counterfeit naira note from original.

A comprehensive collection of foundational Deep Learning projects and experiments, covering manual implementation of the Perceptron and Backpropagation (NumPy) to advanced applications like MLP/CNN image classifiers and LSTM time-series prediction (Keras/TensorFlow).

Flatiron Capstone; Computer Vision model to identify tablets and capsules in pursuit of improving medication safety. Modeled with densely-connected, convolutional neural network, and CNN with transfer learning (VGG16) models;

CNN for number recognition of 28x28 images from MNIST dataset