-
Enhanced V1 Identifies Fakes better V8 Identifies human faces better
-
Update README.md
-
Merge branch 'main' of https://huggingface.co/recklessforlife/Real_Vs_Placeholder_Front_Face
-
Create Test.py
-
Update README.md
-
First Commit
-
Update README.md
-
Update README.md
-
Update README.md
# Model Card: Face Authenticity Classifier
## Model Overview
**Model Name:** Real_vs_Placeholder
**Model Type:** Convolutional Neural Network for Binary Classification
**Task:** Real vs Placeholder Face Detection
**Framework:** PyTorch
**Input Resolution:** 224×224×3 RGB images
**Output:** Binary classification (Real=1, Fake=0)
## Model Architecture
### Network Structure
The model employs a three-block convolutional architecture with progressive feature extraction:
**Feature Extraction Blocks:**
- **Block 1:** 128 filters (224×224 → 112×112)
- **Block 2:** 256 filters (112×112 → 56×56)
- **Block 3:** 512 filters (56×56 → 28×28)
**Each Block Contains:**
- Two 3×3 convolutional layers with same padding
- Batch Normalization after each convolution
- ReLU activation functions
- 2×2 Max Pooling for downsampling
- Dropout (30%) for regularization
**Classification Head:**
- Adaptive Global Average Pooling (7×7 output)
- Fully Connected Layer 1: 25,088 → 1,024 neurons
- Fully Connec… -
initial commit
