File size: 5,862 Bytes

---
license: mit
base_model:
- Modotte/AIRealNet
pipeline_tag: image-classification
library_name: transformers
tags:
- safety
- Modotte
- AIRealNet
- SoTA
---
# Modotte

<p align="center">
  <img 
    src="https://cdn-uploads.huggingface.co/production/uploads/677fcdf29b9a9863eba3f29f/T-Twoka4G5_Qfcqlka4l8.png" 
    alt="AIRealNet Banner" 
    width="90%" 
    style="border-radius:15px;"
  />
</p>

---

> This is an upgraded version of AIRealNet. We have improved its performance on social media images of different qualities, making it a great upgrade from its previous version. 

## Overview

In an era of rapidly advancing AI-generated imagery, deepfakes, and synthetic media, the need for reliable detection tools has never been higher. **AIRealNet-2** is a binary image classifier explicitly designed to distinguish **AI-generated images** from **real human photographs**. This model is optimized to detect conventional AI-generated content while adhering to strict privacy standards—avoiding personal or sensitive images.

* **Class 0:** AI-generated image
* **Class 1:** Real human image

By leveraging the robust **SwinV2 Tiny** architecture as its backbone, AIRealNet achieves a high degree of accuracy while remaining lightweight enough for practical deployment.

---

## Key Features

1. **High Accuracy on Public Datasets:** Model is trained on data from social media 
   
2. **Balanced Training Split:**
   The dataset contains a balanced number of AI-generated and real images, ensuring unbiased training and minimizing class imbalance issues.

   * **AI-Generated:** 50%
   * **Human-Images:** 50%

4. **Ethical Design:**
   No personal photos were included, even if edited or AI-modified, respecting privacy and ethical AI principles.

5. **Fast and Scalable:**
   Based on a transformer vision model, AIRealNet can be deployed efficiently in both research and production environments.

---

## Training Data

* **Dataset:** Collection of images from differnt plaforms like instagram, x.
* **Size:** 98k images (balanced between AI and human)
* **Split:** Used the train split for fine-tuning; validation performed on a separate balanced subset.
* **Notes:** Images sourced from public datasets and AI generation tools. Edited personal photos were intentionally excluded.

---

## Limitations

While AIRealNet performs exceptionally well on typical AI-generated images, users should note:

1. **Subtle Edits:** The model struggles with nano-scale edits or ultra-precise modifications, like “nano banana” edits.
2. **Edited Personal Images(over precise):** Images of real people that have been AI-modified are **not detected**, aligning with privacy and ethical guidelines.
3. **Domain Generalization:** Performance may vary on images from completely unseen AI generators or extremely unconventional content.

---

## Performance Metrics

> Metrics, chosen to illustrate stable performance after fine-tuning.

<p align="center">
  <img 
    src="https://cdn-uploads.huggingface.co/production/uploads/677fcdf29b9a9863eba3f29f/Z7rfOLgZuxDn96ss25bRL.png" 
    alt="AIRealNet Banner" 
    width="90%" 
    style="border-radius:15px;"
  />
</p>

---

## Demo and Usage

1. **Installing dependencies**

```python
pip install -U transformers
```
2. **Loading and running a demo**

```python
from transformers import pipeline

pipe = pipeline("image-classification", model="Modotte/AIRealNet-2")
pipe("https://cdn-uploads.huggingface.co/production/uploads/677fcdf29b9a9863eba3f29f/rYzxmRIFgqg2_EfEFRi89.jpeg")# example image
```
# Demo

* **Given Image**: This image is actually real, but the older model would classify it as fake because it was not trained on varied social media platform compressions.

<p align="center">
  <img 
    src="https://cdn-uploads.huggingface.co/production/uploads/677fcdf29b9a9863eba3f29f/rYzxmRIFgqg2_EfEFRi89.jpeg" 
    alt="AIRealNet Banner" 
    width="90%" 
    style="border-radius:15px;"
  />
</p>

* **New Model Output**

```bash
[{'label': 'real', 'score': 0.9999173879623413},
 {'label': 'artificial', 'score': 8.256531145889312e-05}]
```


---

## Intended Use

* Detect AI-generated imagery on social media, research publications, and digital media platforms.
* Assist content moderators, researchers, and fact-checkers in identifying synthetic media.
* **Not intended** for legal verification without human corroboration.

---

## Ethical Considerations

* **Privacy-first Approach:** Personal photos, even if AI-edited, were excluded.
* **Responsible Deployment:** Users should combine model predictions with human review to avoid false positives or negatives.
* **Transparency:** The model card openly communicates its limitations and dataset design to prevent misuse.

---

## How It Works

1. Images are preprocessed and resized to `256x256`.
2. Features are extracted using the **SwinV2 Tiny** vision transformer backbone.
3. A binary classification head outputs probabilities for AI-generated vs real human images.
4. Predictions are interpreted as class 0 (AI) or class 1 (Human).

---

## Future Work

Future iterations aim to:

* Improve detection of subtle AI-generated edits and “nano banana” modifications.
* Expand training data with diverse AI generators to enhance generalization.
* Explore multi-modal detection capabilities (e.g., video, metadata, and image combined).

---

### Citation
```bibtex
@misc{Modotte_AIRealNet_2025,
    title={AIRealNet-2: A Fine-Tuned Vision Transformer for Detecting AI-Generated vs Real Human Images},
    author={Parvesh Rawal},
    publisher={Hugging Face},
    year={2025},
    url={https://huggingface.co/Modotte/AIRealNet}
}
```

## References

* Microsoft SwinV2 Tiny: [https://github.com/microsoft/Swin-Transformer](https://github.com/microsoft/Swin-Transformer)
* Parveshiiii/AI-vs-Real dataset (subset): Open-sourced by our team member

---