Update app.py

app.py

@@ -7,74 +7,74 @@ from tensorflow.keras.applications.xception import preprocess_input as xcp_pre
 from tensorflow.keras.applications.efficientnet import preprocess_input as eff_pre
 from huggingface_hub import hf_hub_download
 
-
-# Load models
+# Download models from Hugging Face
 xcp_path = hf_hub_download(repo_id="Zeyadd-Mostaffa/deepfake-image-detector_final", filename="xception_model.h5")
 eff_path = hf_hub_download(repo_id="Zeyadd-Mostaffa/deepfake-image-detector_final", filename="efficientnet_model.h5")
 xcp_model = load_model(xcp_path)
 eff_model = load_model(eff_path)
 
-
-# Load face detector
+# Load MTCNN detector
 detector = MTCNN()
 
-def expand_box(x, y, w, h, scale=1.5, img_shape=None):
-    """Expand face bounding box with margin."""
-    cx, cy = x + w // 2, y + h // 2
-    new_w, new_h = int(w * scale), int(h * scale)
-    x1 = max(0, cx - new_w // 2)
-    y1 = max(0, cy - new_h // 2)
-    x2 = min(img_shape[1], cx + new_w // 2)
-    y2 = min(img_shape[0], cy + new_h // 2)
-    return x1, y1, x2, y2
+# Parameters
+MIN_FACE_SIZE = 100      # Skip small distant faces
+MIN_CONFIDENCE = 0.97    # MTCNN face confidence
+AVG_THRESHOLD = 0.41     # Decision boundary
 
 def predict(image):
     faces = detector.detect_faces(image)
     if not faces:
         return "No face detected", image
 
-    output_image = image.copy()
+    output = image.copy()
     results = []
 
     for idx, face in enumerate(faces):
+        if face['confidence'] < MIN_CONFIDENCE:
+            continue
+
         x, y, w, h = face['box']
+        if w < MIN_FACE_SIZE or h < MIN_FACE_SIZE:
+            continue
 
-        # Add 20% margin while staying inside bounds
+        # Expand face bounding box by margin
         margin = 0.2
         img_h, img_w = image.shape[:2]
         x = max(0, int(x - w * margin))
         y = max(0, int(y - h * margin))
-        w = int(w * (1 + 2 * margin))
-        h = int(h * (1 + 2 * margin))
-        x2 = min(img_w, x + w)
-        y2 = min(img_h, y + h)
-
-        face_img = image[y:y2, x:x2]
-
-        # Resize + preprocess
-        face_xcp = cv2.resize(face_img, (299, 299))
-        face_eff = cv2.resize(face_img, (224, 224))
-        xcp_tensor = xcp_pre(face_xcp.astype(np.float32))[np.newaxis, ...]
-        eff_tensor = eff_pre(face_eff.astype(np.float32))[np.newaxis, ...]
-
-        # Predictions
-        pred_xcp = xcp_model.predict(xcp_tensor, verbose=0).flatten()[0]
-        pred_eff = eff_model.predict(eff_tensor, verbose=0).flatten()[0]
-        avg = (pred_xcp + pred_eff) / 2
-
-        label = "Real" if avg > 0.41 else "Fake"
+        x2 = min(img_w, int(x + w * (1 + 2 * margin)))
+        y2 = min(img_h, int(y + h * (1 + 2 * margin)))
+
+        face_crop = image[y:y2, x:x2]
+        if face_crop.size == 0:
+            continue
+
+        # Resize and preprocess for both models
+        xcp_input = cv2.resize(face_crop, (299, 299)).astype(np.float32)
+        eff_input = cv2.resize(face_crop, (224, 224)).astype(np.float32)
+        xcp_tensor = xcp_pre(xcp_input)[np.newaxis, ...]
+        eff_tensor = eff_pre(eff_input)[np.newaxis, ...]
+
+        # Get predictions
+        xcp_score = xcp_model.predict(xcp_tensor, verbose=0).flatten()[0]
+        eff_score = eff_model.predict(eff_tensor, verbose=0).flatten()[0]
+        avg_score = (xcp_score + eff_score) / 2
+        label = "Real" if avg_score > AVG_THRESHOLD else "Fake"
         color = (0, 255, 0) if label == "Real" else (0, 0, 255)
 
         # Annotate image
-        cv2.rectangle(output_image, (x, y), (x2, y2), color, 2)
-        cv2.putText(output_image, f"{label} ({avg:.2f})", (x, y - 10),
+        cv2.rectangle(output, (x, y), (x2, y2), color, 2)
+        cv2.putText(output, f"{label} ({avg_score:.2f})", (x, y - 10),
                     cv2.FONT_HERSHEY_SIMPLEX, 0.6, color, 2)
 
-        results.append(f"Face {idx+1}: {label} (Avg: {avg:.3f}, XCP: {pred_xcp:.3f}, EFF: {pred_eff:.3f})")
+        results.append(f"Face {idx+1}: {label} (Avg: {avg_score:.3f}, XCP: {xcp_score:.3f}, EFF: {eff_score:.3f})")
 
-    return "\n".join(results), output_image
+    if not results:
+        return "No clear or confident face detected", image
 
-# Gradio Interface
+    return "\n".join(results), output
+
+# Gradio UI
 interface = gr.Interface(
     fn=predict,
     inputs=gr.Image(type="numpy", label="Upload Image"),
@@ -83,7 +83,8 @@ interface = gr.Interface(
         gr.Image(type="numpy", label="Annotated Image"),
     ],
     title="Deepfake Detector (Multi-Face Ensemble)",
-    description="Detects all faces in an image and classifies each one as real or fake using Xception and EfficientNetB4 ensemble.",
+    description="Detects and classifies confident faces as Real or Fake using Xception + EfficientNetB4.",
 )
 
 interface.launch()
+