Skip to content

Latest commit

 

History

History
42 lines (22 loc) · 918 Bytes

README.md

File metadata and controls

42 lines (22 loc) · 918 Bytes

COM_COM_000166-Deepfake-Detection

Please find further details about MesoNet and YOLOv5 in the respective folders.

For train/val data please contact FaceForensics++ https://github.com/ondyari/FaceForensics#access

Process Flow

Data PreProcessing

Data_PreProcessing

Voting Mechanism

Voting

RESULTS

Accuracy_all5

Accuracy_individual

We have publised the reults of YOLOv5 on the here - http://kaldir.vc.in.tum.de/faceforensics_benchmark/

MesoNet Benchmarking

MesoNet

YOLOv5 Benchmarking

YOLOv5

YOLOv5 with Voting Mechanism Benchmarking

YOLOv5_Voting

Front End

In the YOLOv5 directory, one can simply execute python3 app.py to run flask server and launch site on localhost.

Front_End