Looking for AI-edit detection and tamper heatmaps? See Car Damage Photo Integrity v9 for image forgery localization in car-damage photos.

This repository retains its original checkpoint and inference instructions.

Claim Photo Integrity Detector v2

v2 of the claim-photo integrity detector. Continued fine-tuning of v1 with a much broader "real" diet: Open Images V7 everyday photos, KonIQ-10k authentically-distorted photos (blur / noise / bad light), and real phone photos - plus training-time augmentation (JPEG recompression, noise, blur, brightness/contrast jitter) so everyday phone photography is recognized as real instead of flagged AI-generated.

Usage

from transformers import pipeline
pipe = pipeline("image-classification",
                model="obuladinnesai/claim-photo-integrity-detector-v2")
pipe("path/to/claim_photo.jpg")

Limitations

• A triage signal for human review, not proof of fraud.
• Accuracy degrades on generators newer than the training data.