Create app.py

app.py

@@ -0,0 +1,85 @@
+from transformers import AutoImageProcessor, SiglipForImageClassification
+from PIL import Image
+import torch
+import cv2
+import os
+import gradio as gr
+
+# Load model and processor
+model_name = "prithivMLmods/deepfake-detector-model-v1"
+model = SiglipForImageClassification.from_pretrained(model_name)
+processor = AutoImageProcessor.from_pretrained(model_name)
+
+# Updated label mapping
+id2label = {
+    "0": "fake",
+    "1": "real"
+}
+
+def classify_image(image):
+    image = Image.fromarray(image).convert("RGB")
+    inputs = processor(images=image, return_tensors="pt")
+
+    with torch.no_grad():
+        outputs = model(**inputs)
+        logits = outputs.logits
+        probs = torch.nn.functional.softmax(logits, dim=1).squeeze().tolist()
+
+    prediction = {
+        id2label[str(i)]: round(probs[i], 3) for i in range(len(probs))
+    }
+
+    return prediction
+
+def sliceFrames(cap):
+    frame_count = 0
+    frames = []
+
+    while True:
+        ret, frame = cap.read()
+        if not ret:
+            break
+        # Save every 10th frame
+        if frame_count % 10 == 0:
+            frames.append(frame)
+        frame_count += 1
+
+    cap.release()
+    return frames
+
+def classify_video(video_path):
+    cap = cv2.VideoCapture(video_path)
+    if not cap.isOpened():
+        return {"error": "Could not open video."}
+
+    frames = sliceFrames(cap)
+    totalfake = 0
+    totalreal = 0
+
+    for frame in frames:
+        prediction = classify_image(frame)
+        totalfake += prediction["fake"]
+        totalreal += prediction["real"]
+
+    avg_fake = totalfake / len(frames) if frames else 0
+    avg_real = totalreal / len(frames) if frames else 0
+
+    return {
+        "average_fake": round(avg_fake, 3),
+        "average_real": round(avg_real, 3),
+    }
+
+# Gradio Interface
+def gradio_interface(video_file):
+    return classify_video(video_file)
+
+iface = gr.Interface(
+    fn=gradio_interface,
+    inputs=gr.Video(label="Upload a video"),
+    outputs=gr.JSON(label="Prediction"),
+    title="Deepfake Detector",
+    description="Upload a video to check if it's real or fake."
+)
+
+if __name__ == "__main__":
+    iface.launch()