|
|
--- |
|
|
license: apache-2.0 |
|
|
pipeline_tag: image-classification |
|
|
library_name: transformers |
|
|
tags: |
|
|
- deep-fake |
|
|
- detection |
|
|
- Image |
|
|
- SigLIP2 |
|
|
base_model: |
|
|
- google/siglip2-base-patch16-512 |
|
|
datasets: |
|
|
- prithivMLmods/OpenDeepfake-Preview |
|
|
language: |
|
|
- en |
|
|
--- |
|
|
|
|
|
 |
|
|
|
|
|
# deepfake-detector-model-v1 |
|
|
|
|
|
> `deepfake-detector-model-v1` is a vision-language encoder model fine-tuned from google/siglip-base-patch16-512 for binary deepfake image classification. It is trained to detect whether an image is real or generated using synthetic media techniques. The model uses the `SiglipForImageClassification` architecture. |
|
|
|
|
|
> [!warning] |
|
|
Experimental |
|
|
|
|
|
```py |
|
|
Classification Report: |
|
|
precision recall f1-score support |
|
|
|
|
|
Fake 0.9718 0.9155 0.9428 10000 |
|
|
Real 0.9201 0.9734 0.9460 9999 |
|
|
|
|
|
accuracy 0.9444 19999 |
|
|
macro avg 0.9459 0.9444 0.9444 19999 |
|
|
weighted avg 0.9459 0.9444 0.9444 19999 |
|
|
``` |
|
|
|
|
|
 |
|
|
|
|
|
--- |
|
|
|
|
|
## Label Space: 2 Classes |
|
|
|
|
|
The model classifies an image as one of the following: |
|
|
|
|
|
``` |
|
|
Class 0: fake |
|
|
Class 1: real |
|
|
``` |
|
|
|
|
|
--- |
|
|
|
|
|
## Install Dependencies |
|
|
|
|
|
```bash |
|
|
pip install -q transformers torch pillow gradio hf_xet |
|
|
``` |
|
|
|
|
|
--- |
|
|
|
|
|
## Inference Code |
|
|
|
|
|
```python |
|
|
import gradio as gr |
|
|
from transformers import AutoImageProcessor, SiglipForImageClassification |
|
|
from PIL import Image |
|
|
import torch |
|
|
|
|
|
# Load model and processor |
|
|
model_name = "prithivMLmods/deepfake-detector-model-v1" |
|
|
model = SiglipForImageClassification.from_pretrained(model_name) |
|
|
processor = AutoImageProcessor.from_pretrained(model_name) |
|
|
|
|
|
# Updated label mapping |
|
|
id2label = { |
|
|
"0": "fake", |
|
|
"1": "real" |
|
|
} |
|
|
|
|
|
def classify_image(image): |
|
|
image = Image.fromarray(image).convert("RGB") |
|
|
inputs = processor(images=image, return_tensors="pt") |
|
|
|
|
|
with torch.no_grad(): |
|
|
outputs = model(**inputs) |
|
|
logits = outputs.logits |
|
|
probs = torch.nn.functional.softmax(logits, dim=1).squeeze().tolist() |
|
|
|
|
|
prediction = { |
|
|
id2label[str(i)]: round(probs[i], 3) for i in range(len(probs)) |
|
|
} |
|
|
|
|
|
return prediction |
|
|
|
|
|
# Gradio Interface |
|
|
iface = gr.Interface( |
|
|
fn=classify_image, |
|
|
inputs=gr.Image(type="numpy"), |
|
|
outputs=gr.Label(num_top_classes=2, label="Deepfake Classification"), |
|
|
title="deepfake-detector-model", |
|
|
description="Upload an image to classify whether it is real or fake using a deepfake detection model." |
|
|
) |
|
|
|
|
|
if __name__ == "__main__": |
|
|
iface.launch() |
|
|
``` |
|
|
|
|
|
--- |
|
|
|
|
|
## Intended Use |
|
|
|
|
|
`deepfake-detector-model` is designed for: |
|
|
|
|
|
* **Deepfake Detection** – Accurately identify fake images generated by AI. |
|
|
* **Media Authentication** – Verify the authenticity of digital visual content. |
|
|
* **Content Moderation** – Assist in filtering synthetic media in online platforms. |
|
|
* **Forensic Analysis** – Support digital forensics by detecting manipulated visual data. |
|
|
* **Security Applications** – Integrate into surveillance systems for authenticity verification. |
