M

Mirage Photo Classifier

Developed by prithivMLmods
An image classification model fine-tuned based on SigLIP2, used to detect whether an image is real or AI-generated
Downloads 16
Release Time : 4/4/2025

Model Overview

The Mirage Photo Classifier is a vision-language encoding model fine-tuned from google/siglip2-base-patch16-224, specifically designed for binary classification tasks in image authenticity detection. This model adopts the SiglipForImageClassification architecture to determine whether an image is real or AI-generated (forged).

Model Features

High-Precision Classification
Achieves 94.64% accuracy on the test set with an F1 score of 0.9463
Dual-Category Detection
Capable of distinguishing between real photographs and AI-generated images
Based on SigLIP2 Architecture
Utilizes google/siglip2-base-patch16-224 as the base model for fine-tuning

Model Capabilities

Image Authenticity Detection
AI-Generated Image Recognition
Binary Image Classification

Use Cases

Content Moderation
Social Media Content Moderation
Identifying AI-generated images on social media platforms
Helps platforms flag generated content
Digital Forensics
Image Authenticity Investigation
Assisting professionals in investigating image authenticity
Provides probability assessments of image authenticity
Data Validation
Training Data Cleaning
Validating training data for other AI models
Improves dataset quality
Featured Recommended AI Models
AIbase
Empowering the Future, Your AI Solution Knowledge Base
© 2025AIbase