DeepFake-o-meter

Task: Real or Forged Image Detection

Analysis Date: 2025-09-24 00:50:47.794705

Information:
Resolution: 327 x 243 | Codec: PNG | File Size: 156.98KB

Analytic Name: HIFI

Description: we first represent forgery attributes of a manipulated image with multiple labels at different levels. Then we perform fine-grained classification at these levels using the hierarchical dependency between them.

Reference: https://arxiv.org/pdf/2303.17111.pdf

Code Link: https://github.com/CHELSEA234/HiFi_IFDL

🧪 Fake Probability: 65.5%

▲

65.5%

AI-generated Likelihood

Scope: Testing if an image is real or forged (fully synthesized or partially manipulated).

Result Details