Updated

- URL as Image Source

app.py

@@ -5,11 +5,13 @@ from PIL import Image
 from torchvision import transforms
 import json
 import cv2
+import requests
+from io import BytesIO
 from src.model import SwinTransformerMultiLabel  # Import from src folder
 
 # Title and description
 st.title("STAR Multi-Label Classifier")
-st.write("Upload an image to classify and blur sensitive areas.")
+st.write("Upload an image or provide a URL to classify and blur sensitive areas.")
 
 # Load class labels from JSON
 label_file = "data/labels.json"
@@ -36,7 +38,7 @@ transform = transforms.Compose([
     transforms.Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225])
 ])
 
-# Function to detect sensitive parts using OpenCV
+# Function to detect sensitive areas using OpenCV
 def detect_sensitive_areas(image):
     """Detect sensitive areas (breasts, vagina, penis) using OpenCV."""
     image_cv = np.array(image)  # Convert PIL to NumPy
@@ -68,11 +70,27 @@ def blur_sensitive_parts(image, mask, blur_intensity=25):
 
     return Image.fromarray(result)  # Convert back to PIL Image
 
-# Upload image
-uploaded_file = st.file_uploader("Choose an image...", type=["jpg", "jpeg", "png"])
-if uploaded_file is not None:
-    image = Image.open(uploaded_file).convert("RGB")
+# UI for image input: Upload or URL
+option = st.radio("Choose an input method:", ("Upload Image", "Enter Image URL"))
 
+image = None  # Placeholder for the image
+
+if option == "Upload Image":
+    uploaded_file = st.file_uploader("Choose an image...", type=["jpg", "jpeg", "png"])
+    if uploaded_file is not None:
+        image = Image.open(uploaded_file).convert("RGB")
+
+elif option == "Enter Image URL":
+    image_url = st.text_input("Enter image URL:")
+    if image_url:
+        try:
+            response = requests.get(image_url)
+            image = Image.open(BytesIO(response.content)).convert("RGB")
+        except Exception as e:
+            st.error(f"Error fetching image: {e}")
+
+# Process the image if provided
+if image:
     # Preprocess image for model
     img_tensor = transform(image).unsqueeze(0)