Update README.md

README.md

@@ -1,22 +1,24 @@
-# Model Title
-BioSafe - Facial Liveness Detection for Anti-Spoofing.
+---
+title: Facial Liveness Detection
+emoji: 🚫
+colorFrom: red
+colorTo: green
+sdk: gradio
+sdk_version: 3.14.0
+app_file: app.py
+pinned: false
+---
 
-## Model Description
-This model is designed to detect and differentiate between genuine human faces and spoofed faces in images. It's aimed at enhancing security in systems where facial recognition is used by identifying attempts to trick the system using photographs, videos, or masks.
-
-## Model Architecture
-The model is based on the ResNet-50 architecture, a 50-layer deep convolutional neural network. It has been fine-tuned for the specific task of facial liveness detection. The network employs a combination of convolutional layers, pooling layers, and fully connected layers, with a final sigmoid activation function for binary classification (genuine or spoofed).
 
-## Training Data
-The model was trained on a dataset consisting of approximately 17,500 frames, with a mix of genuine and spoofed facial images. The spoofed images include various types of attacks like photographs and video replays.
+# Facial Liveness Detection
+This Gradio app demonstrates a facial liveness detection model that can differentiate between real human faces and spoofed faces.
 
-## Training Procedure
-The model was trained using transfer learning, where the ResNet-50 base was initialized with ImageNet weights. The top layers were customized and trained on our dataset. We used a binary cross-entropy loss function and the Adam optimizer. Training was performed with early stopping to prevent overfitting.
+## Model Description
+This model is trained to determine the authenticity of faces in images, aiming to prevent spoofing attacks in systems where facial recognition is a part of the security process.
 
-## Intended Use
-This model is intended for use in systems where facial recognition is a component of authentication or security processes. It can help in distinguishing between actual users and spoofed attempts, enhancing overall system security.
+## How to Use the App
+1. Use the webcam to capture your image or upload an image using the interface.
+2. The app will process the image and display it with a label indicating whether the face is real or spoofed.
 
 ## Limitations and Bias
-The model's effectiveness may vary across different demographic groups depending on the representation in the training data.
-Performance might be lower in real-world scenarios with varying lighting, angles, and image quality compared to the controlled conditions of the training data.
-The model may struggle with very sophisticated spoofing techniques not represented in the training set.
+The model's performance may vary across different demographic groups and is dependent on the quality and diversity of the training data.