Update app.py

app.py

@@ -1,38 +1,168 @@
 import gradio as gr
-from stress_detection import detect_stress_from_image  # Import your function
+import cv2
+import mediapipe as mp
+import numpy as np
+from stress_detection import *  # Import your entire code
+
+# Initialize detectors (create once, reuse for all images)
+au01 = AU01Detector()
+au04 = AU04Detector()
+au06 = AU06Detector()
+au07 = AU07Detector()
+au12 = AU12Detector()
+au14 = AU14Detector()
+au17 = AU17Detector()
+au23 = AU23Detector()
+au24 = AU24Detector()
+au26 = AU26Detector()
+
+detectors = [au01, au04, au06, au07, au12, au14, au17, au23, au24, au26]
+
+# Initialize MediaPipe
+mp_face_mesh = mp.solutions.face_mesh
+face_mesh = mp_face_mesh.FaceMesh(
+    min_detection_confidence=0.5,
+    min_tracking_confidence=0.5,
+    refine_landmarks=True
+)
+
+def detect_stress_from_single_image(image):
+    """
+    Process a single image and return stress prediction
+    
+    Args:
+        image: numpy array (BGR from Gradio)
+    
+    Returns:
+        dict: Stress classification with probabilities
+    """
+    if image is None:
+        return {"Error": "No image provided"}
+    
+    try:
+        # Get image dimensions
+        frame_height, frame_width = image.shape[:2]
+        
+        # Convert to RGB for MediaPipe
+        rgb_frame = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
+        
+        # Process with MediaPipe
+        results = face_mesh.process(rgb_frame)
+        
+        if not results.multi_face_landmarks:
+            return {"Error": "No face detected in image"}
+        
+        landmarks = results.multi_face_landmarks[0].landmark
+        
+        # Detect all 10 AUs
+        au01_active, au01_intensity = au01.detect(landmarks, frame_width, frame_height)
+        au04_active, au04_intensity = au04.detect(landmarks, frame_width, frame_height)
+        au06_active, au06_intensity = au06.detect(landmarks, frame_width, frame_height)
+        au07_active, au07_intensity = au07.detect(landmarks, frame_width, frame_height)
+        au12_active, au12_intensity = au12.detect(landmarks, frame_width, frame_height)
+        au14_active, au14_intensity = au14.detect(landmarks, frame_width, frame_height)
+        au17_active, au17_intensity = au17.detect(landmarks, frame_width, frame_height)
+        au23_active, au23_intensity = au23.detect(landmarks, frame_width, frame_height)
+        au24_active, au24_intensity = au24.detect(landmarks, frame_width, frame_height)
+        au26_active, au26_intensity = au26.detect(landmarks, frame_width, frame_height)
+        
+        # Calculate simple stress score from single frame
+        stress_aus_intensity = (au01_intensity + au04_intensity + au07_intensity + 
+                                au17_intensity + au23_intensity + au24_intensity) / 6
+        
+        positive_aus_intensity = (au06_intensity + au12_intensity + au14_intensity) / 3
+        
+        # Simple scoring for single frame
+        stress_score = stress_aus_intensity - (positive_aus_intensity * 0.3)
+        stress_score = min(100, max(0, stress_score))
+        
+        # Classification
+        if stress_score < 30:
+            return {
+                "Not Stressed 😊": float(100 - stress_score) / 100,
+                "Stressed 😰": float(stress_score) / 100
+            }
+        else:
+            return {
+                "Stressed 😰": float(stress_score) / 100,
+                "Not Stressed 😊": float(100 - stress_score) / 100
+            }
+        
+    except Exception as e:
+        return {"Error": f"Processing failed: {str(e)}"}
+
 
 # Create Gradio interface
-with gr.Blocks(title="Stress Detection System") as demo:
+with gr.Blocks(title="Stress Detection System", theme=gr.themes.Soft()) as demo:
     gr.Markdown("# 🧠 Stress Detection System")
-    gr.Markdown("### AI-Powered Real-Time Stress Analysis")
+    gr.Markdown("### AI-Powered Facial Expression Analysis using 10 Action Units (FACS)")
+    gr.Markdown("*Based on Facial Action Coding System - Ekman & Friesen*")
     
     with gr.Row():
         with gr.Column():
             image_input = gr.Image(
                 sources=["webcam"],
                 type="numpy",
-                label="📹 Webcam"
+                label="📹 Capture Your Face"
             )
-            analyze_btn = gr.Button("🔍 Analyze Stress", variant="primary", size="lg")
-        
+            analyze_btn = gr.Button("🔍 Analyze Stress Level", variant="primary", size="lg")
+            
+            gr.Markdown("""
+            ### 📝 How to use:
+            1. Click the **webcam icon** above
+            2. Allow camera access
+            3. Position your face clearly in frame
+            4. Click **Analyze Stress Level**
+            """)
+            
         with gr.Column():
             output_label = gr.Label(
                 num_top_classes=2,
-                label="📊 Results"
+                label="📊 Stress Analysis Results"
             )
+            
+            gr.Markdown("""
+            ### 🔬 Action Units Detected:
+            **Stress Indicators:**
+            - AU01: Inner Brow Raise
+            - AU04: Brow Lowerer  
+            - AU07: Lid Tightener
+            - AU17: Chin Raiser
+            - AU23: Lip Tightener
+            - AU24: Lip Pressor
+            
+            **Positive Indicators:**
+            - AU06: Cheek Raiser
+            - AU12: Lip Corner Puller
+            - AU14: Dimpler
+            - AU26: Jaw Drop
+            """)
     
-    gr.Markdown("### 📝 How to use:")
-    gr.Markdown("1. Click the webcam icon to enable your camera")
-    gr.Markdown("2. Position your face in the frame")
-    gr.Markdown("3. Click **Analyze Stress** to get prediction")
-    
-    # Connect button to your function
+    # Connect button
     analyze_btn.click(
-        fn=detect_stress_from_image,  # Your function name
+        fn=detect_stress_from_single_image,
         inputs=image_input,
         outputs=output_label
     )
+    
+    gr.Markdown("""
+    ---
+    ### ℹ️ About
+    This system uses **10 Facial Action Units** based on the Facial Action Coding System (FACS) 
+    to detect stress through micro-expressions. Developed under the guidance of **Prof. Anup Nandy**.
+    """)
 
 # Launch
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()
+```
+
+---
+
+## Final File Structure in Your Space:
+```
+your-space/
+├── app.py                  # ← Gradio interface (code above)
+├── stress_detection.py     # ← Your complete code (paste as-is)
+├── requirements.txt        # ← 6 packages listed above
+└── README.md              # ← Update with proper metadata