added app.py

app.py

@@ -1,28 +1,41 @@
-import av
 import streamlit as st
 import cv2  # OpenCV for image processing
-import numpy as np  # <--- REQUIRED: Must be in requirements.txt (pip install av)
+import numpy as np
+import av  # REQUIRED for VideoProcessorBase
 from streamlit_webrtc import webrtc_streamer, VideoProcessorBase, WebRtcMode 
 
-# --- PLACEHOLDER IMPORTS (Adjust as needed) ---
-# ... 
+# --- PLACEHOLDER IMPORTS (UNCOMMENT/ADJUST AS NEEDED) ---
+# NOTE: Make sure these core libraries are in your requirements.txt
+# import deepface 
+# from deepface import DeepFace # Example import if using deepface
+# from src.detect import detect_faces
+# from src.recognize import recognize_face
 
 # --- CONFIGURATION ---
 RECOGNITION_THRESHOLD = 0.6
 FRAME_SKIP = 3
 
-# --- CACHE THE MODEL/FACTORY (CRITICAL FIX FOR THREADING) ---
-# This ensures the processor and its threads are initialized only ONCE.
+
+# --- CRITICAL FIX 1: CACHE THE HEAVY MODELS SEPARATELY ---
+@st.cache_resource
+def load_deepface_models():
+    """Loads all heavy models (DeepFace, etc.) only once, safely outside the thread."""
+    # NOTE: Replace 'return "Loaded Models"' with your actual model loading logic
+    # Example: return DeepFace.build_model("VGG-Face")
+    return "Loaded Models" 
+
+
+# --- CRITICAL FIX 2: CACHE THE FACTORY ---
 @st.cache_resource
 def get_face_processor_factory():
-    """Returns the processor class, ensuring it is cached."""
+    """Returns the processor class, ensuring its initialization is thread-safe."""
     
     class FaceRecognitionProcessor(VideoProcessorBase):
         """Processes video frames using the standard VideoProcessorBase."""
         def __init__(self):
-            # Initialize models once here (safely cached by @st.cache_resource)
+            # Load models from the cached function, reducing memory strain on thread start
+            self.models = load_deepface_models() 
             self.frame_count = 0
-            # Example: self.detection_model = load_your_model()
             
         def recv(self, frame: av.VideoFrame) -> av.VideoFrame:
             # Convert frame from av.VideoFrame to numpy array (BGR)
@@ -34,6 +47,7 @@ def get_face_processor_factory():
             
             # --- Your Face Detection and Recognition Logic Goes Here ---
             h, w, _ = img.shape
+            
             # Placeholder drawing logic:
             x, y, w_box, h_box = w//4, h//4, w//2, h//2 
             cv2.rectangle(img, (x, y), (x + w_box, y + h_box), (0, 0, 255), 2)
@@ -51,14 +65,16 @@ def main():
     st.title("Smart Office Face Recognition System 📸")
     st.sidebar.title("Configuration")
 
-    # ... (Sidebar control for threshold) ...
+    RECOGNITION_THRESHOLD = st.sidebar.slider(
+        "Recognition Threshold", min_value=0.0, max_value=1.0, value=0.6, step=0.05
+    )
 
     # --- FINAL WEBRTC STREAMER CALL ---
     webrtc_streamer(
         key="face-recognition-stream-final-cache",  
         mode=WebRtcMode.SENDRECV,
         
-        # --- Enhanced STUN/TURN configuration ---
+        # --- Enhanced STUN/TURN configuration to stabilize cloud connection ---
         rtc_configuration={
             "iceServers": [
                 {"urls": ["stun:stun.l.google.com:19302"]},
@@ -76,5 +92,4 @@ def main():
 
 # --- EXECUTION ---
 if __name__ == "__main__":
-    # IMPORTANT: Ensure 'av' is in your requirements.txt
     main()