Skip to content
View Ammuelizabeth's full-sized avatar

Block or report Ammuelizabeth

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Ammuelizabeth/README.md

Hi 👋, I'm Ammu Elizabeth Alexander

AI & Machine Learning Enthusiast • Research-Driven Developer

Computer Vision • Generative AI • Trustworthy & Ethical AI


🌟 About Me

I’m an AI/ML practitioner passionate about transforming research ideas into real-world intelligent systems.
My work blends deep learning, image forensics, and human-centric AI, with a strong focus on clarity, reproducibility, and impact.

  • 🧠 Exploring Deep Learning & Applied AI
  • 🔬 Designing systems aligned with IEEE-style research
  • 🛠️ Building end-to-end ML pipelines (data → model → UI)
  • ✨ Clean code, meaningful experiments, honest evaluation


🚨 Featured Project

DeepReveal is a research-oriented deep learning framework designed to identify pixel-wise AI-generated or manipulated regions within images, rather than only giving a global label.

What makes DeepReveal unique

  • 🧩 Pixel-level localization of AI-generated or tampered content
  • 🙂 Face-aware region analysis using integrated face detection for focused inspection
  • 🎯 Highlights where manipulation occurs, not just whether it exists
  • 🌐 Interactive Flask-based interface for visual inspection
  • 📈 Built with research extensibility and dataset scalability in mind

📌 Applications:
Digital image forensics • Deepfake localization • Media authentication • Misinformation analysis


🔬 Other Projects

🖼️ Deepfake Detection Web Application

MobileNetV3-Large + multi-feature extraction for detecting GAN-generated images, deployed with Streamlit. A CNN-based classification system that determines whether an image is real or fake.
Built using MobileNetV3 and deployed with a Streamlit web interface for real-time inference.


🧰 Tech Stack

Python R SQL

🤖 Machine Learning & DL

PyTorch TensorFlow scikit-learn OpenCV

🛠️ Tools & Platforms

Jupyter Git Streamlit LaTeX


📊 GitHub Analytics

GitHub Stats Top Languages

GitHub Streak


🎯 What I Care About

  • 🔍 Trustworthy & explainable AI
  • 📊 High-quality datasets & preprocessing
  • 🔁 Reproducible ML experiments
  • 📝 Clear documentation & technical writing

🤝 Let’s Connect

💻 GitHub💼 LinkedIn📧 Email


If you find my work useful, consider starring a repository — it truly helps!

Popular repositories Loading

  1. Deepfake-Detection-Web-Application Deepfake-Detection-Web-Application Public

    Developed a real-time web app using Python and Streamlit to detect GAN-generated fake face images. Leveraged a fine-tuned MobileNetV3 model trained on a balanced dataset of 39,000+ images. The syst…

    Jupyter Notebook 4 1

  2. Ammuelizabeth Ammuelizabeth Public

    Config files for my GitHub profile.

    1

  3. DeepReveal DeepReveal Public

    DeepReveal is a deep learning-based image forensics system that performs pixel-level detection of AI-generated or manipulated image regions using U-Net segmentation, PyTorch, OpenCV, and Flask.

    Jupyter Notebook 1

  4. cross-season-image-matching cross-season-image-matching Public

    Deep feature-based visual localization framework using SuperPoint, LightGlue, and homography estimation for robust image matching across extreme seasonal changes.

    Jupyter Notebook