ViT_Deepfake_Detection

Model overview

ViT_Deepfake_Detection is published by Wvolf as an image-classification model. The captured configuration identifies ViTForImageClassification with a vit model type and Transformers support. According to the model card, the model was trained by Rudolf Enyimba for an MSc degree project to detect deepfake images.

Recorded capabilities

Use cases in the source record

• Face-image deepfake screening where a user uploads a face image or selects samples to test predictions, as described in the model card.
• Academic deepfake-detection experiments building on the reported training setup and test accuracy, according to the model card.

Limitations and unknowns

• The reported 98.70% accuracy is a publisher model-card claim without an extracted dataset, split, or independent verification.
• No parameter count was extracted from this record.
• No license value was extracted from this record.
• Provider state is historical snapshot data, not independently refreshed current availability.

Source and provenance

Source: Wvolf/ViT_Deepfake_Detection

Captured: Unknown. Processed: 2026-09-07T19:34:38.867378+00:00.

This model was trained by Rudolf Enyimba in partial fulfillment of the requirements of Solent University for the degree of MSc Artificial Intelligence and Data Science This model was trained to detect deepfake images. The model achieved an accuracy of 98.70% on the test set. Upload a face image or pick from the samples below to test model accuracy