BeyzaTopbas/Facial__Keypoints__Detection

🧠 Facial Keypoints Detection

A Streamlit application that predicts facial keypoints using a trained ResNet model.

📌 What this app does

• —Upload a face image
• —Converts image to grayscale
• —Resizes to 96x96
• —Predicts facial keypoints (eyes, nose, mouth, etc.)
• —Visualizes the keypoints directly on the image

🖼 Example Workflow

1. 1.Upload a front-facing face image
2. 2.Model predicts keypoints
3. 3.Red dots appear on facial landmarks

🏗 Model

• —Architecture: ResNet-based CNN
• —Input shape: 96x96 grayscale
• —Output: (x, y) coordinates for facial landmarks
• —Loss used during training: MSE

🛠 Tech Stack

• —Streamlit
• —TensorFlow / Keras
• —NumPy
• —Matplotlib
• —Pillow

🚀 Run locally

pip install -r requirements.txt
streamlit run src/streamlit_app.py