Spaces:

Wolowolo
/

FSFM-deepfake_diffusion_spoof_face_detection

Running

App Files Files Community

Wolowolo commited on Feb 22

Commit

1671f8a

verified ·

1 Parent(s): b3c9b2c

Update app.py

Browse files

Files changed (1) hide show

app.py +1 -1

app.py CHANGED Viewed

@@ -275,7 +275,7 @@ with gr.Blocks(css=".custom-label { font-weight: bold !important; font-size: 16p
                 "1) Updated <b>[✨Unified-detector_v1] for Unified Physical-Digital Face Attack&Forgery Detection, a vanilla ViT-B/16-224 (FSFM Pre-trained) that could identify Real&Bonafide, Deepfake, Diffduion&AIGC, Spooing&Presentation-attacks facial images or videos </b> ; 2) Provided the selection of the number of video frames (uniformly sampling, more frames are too time-consuming, and we would be grateful if you support us to open paid GPU acceleration); 3) Fixed the errors of V0.1 including loading model and prediction. <br>"
                 "<b>[V0.1]</b> 2024/12-2025/02/21: "
                 "Create this page with basic detectors [DfD-Checkpoint_Fine-tuned_on_FF++, FAS-Checkpoint_Fine-tuned_on_MCIO] that follow the paper implementation. <br> ")
-    gr.Markdown("- Please <b>provide a facial image or video(<100s)</b>, and <b>select the model</b> for detection: <br> <b>[suggest] [✨Unified-detector_v1_Fine-tuned_on_4_classes]</b> <b>a (FSFM Pre-trained) ViT-B/16-224 for Both Real/Deepfake/Diffusion/Spoofing facial images&videos Detection <b> <br> <b>[DfD-Checkpoint_Fine-tuned_on_FF++]</b> for deepfake detection, FSFM ViT-B/16-224 fine-tuned on the FF++_c23 train&val sets (4 manipulations, 32 frames per video) <br> <b>[FAS-Checkpoint_Fine-tuned_on_MCIO]</b> for face anti-spoofing, FSFM ViT-B/16-224 fine-tuned on the MCIO datasets (2 frames per video)")
     with gr.Row():

                 "1) Updated <b>[✨Unified-detector_v1] for Unified Physical-Digital Face Attack&Forgery Detection, a vanilla ViT-B/16-224 (FSFM Pre-trained) that could identify Real&Bonafide, Deepfake, Diffduion&AIGC, Spooing&Presentation-attacks facial images or videos </b> ; 2) Provided the selection of the number of video frames (uniformly sampling, more frames are too time-consuming, and we would be grateful if you support us to open paid GPU acceleration); 3) Fixed the errors of V0.1 including loading model and prediction. <br>"
                 "<b>[V0.1]</b> 2024/12-2025/02/21: "
                 "Create this page with basic detectors [DfD-Checkpoint_Fine-tuned_on_FF++, FAS-Checkpoint_Fine-tuned_on_MCIO] that follow the paper implementation. <br> ")
+    gr.Markdown("- Please <b>provide a facial image or video(<100s)</b>, and <b>select the model</b> for detection: <br> <b>[SUGGEST] [✨Unified-detector_v1_Fine-tuned_on_4_classes]</b> a (FSFM Pre-trained) ViT-B/16-224 for Both Real/Deepfake/Diffusion/Spoofing facial images&videos Detection <br> <b>[DfD-Checkpoint_Fine-tuned_on_FF++]</b> for deepfake detection, FSFM ViT-B/16-224 fine-tuned on the FF++_c23 train&val sets (4 manipulations, 32 frames per video) <br> <b>[FAS-Checkpoint_Fine-tuned_on_MCIO]</b> for face anti-spoofing, FSFM ViT-B/16-224 fine-tuned on the MCIO datasets (2 frames per video)")
     with gr.Row():