Update app.py

app.py

@@ -14,7 +14,7 @@ logging.basicConfig(level=logging.INFO)
 logger = logging.getLogger(__name__)
 
 try:
-    from microexpression import track_microexpressions, get_lip_engagement
+    from microexpression import track_microexpressions
     import mediapipe as mp
 except ImportError as e:
     logger.error(f"Failed to import required modules: {e}")
@@ -154,7 +154,7 @@ class VideoAnalyzer:
             logger.error(f"Error validating video file: {e}")
             raise
     
-    def analyze_video(self, video_file) -> Tuple[str, Dict, Dict]:
+    def analyze_video(self, video_file) -> Tuple[str, Dict]:
         """
         Main video analysis function.
         
@@ -162,7 +162,7 @@ class VideoAnalyzer:
             video_file: Input video file
             
         Returns:
-            Tuple of (result_json, lip_summary, micro_summary)
+            Tuple of (result_json, micro_summary)
         """
         video_path = None
         try:
@@ -186,7 +186,6 @@ class VideoAnalyzer:
             processed_frames = 0
             eye_away_count = 0
             head_turn_count = 0
-            lip_engagement_predictions = []
             emotion_prob_list = []
             stress_score_list = []
             stress_label_list = []
@@ -207,16 +206,6 @@ class VideoAnalyzer:
                             frame, self.face_mesh, self.calibration_ref
                         )
                         
-                        # Lip engagement detection
-                        frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
-                        results = self.face_mesh.process(frame_rgb)
-                        
-                        lip_engagement = "No Face"
-                        if results.multi_face_landmarks:
-                            landmarks = [(lm.x, lm.y) for lm in results.multi_face_landmarks[0].landmark]
-                            lip_engagement = get_lip_engagement(landmarks)
-                        lip_engagement_predictions.append(lip_engagement)
-                        
                         # Update microexpression counts
                         if micro.get("eye_away", False):
                             eye_away_count += 1
@@ -241,7 +230,6 @@ class VideoAnalyzer:
             cap.release()
             
             # Generate summaries
-            lip_summary = self._generate_lip_summary(lip_engagement_predictions)
             micro_summary = self._generate_micro_summary(eye_away_count, head_turn_count, processed_frames)
             emotion_summary = self._generate_emotion_summary(emotion_prob_list)
             stress_summary = self._generate_stress_summary(stress_score_list, stress_label_list)
@@ -254,16 +242,14 @@ class VideoAnalyzer:
                     "processed_frames": processed_frames,
                     "fps": fps
                 },
-                "engagement": lip_summary.get("majority_label", "Unknown"),
                 "stress": stress_summary,
                 "emotion_probabilities": emotion_summary,
                 "confidence_metrics": {
-                    "frames_with_face": len([x for x in lip_engagement_predictions if x != "No Face"]),
                     "total_processed": processed_frames
                 }
             }
             
-            return json.dumps(result, indent=2), lip_summary, micro_summary
+            return json.dumps(result, indent=2), micro_summary
             
         except Exception as e:
             logger.error(f"Error in video analysis: {e}")
@@ -271,7 +257,7 @@ class VideoAnalyzer:
                 "error": str(e),
                 "status": "failed"
             }
-            return json.dumps(error_result, indent=2), {}, {}
+            return json.dumps(error_result, indent=2), {}
         
         finally:
             # Clean up temporary files
@@ -283,20 +269,8 @@ class VideoAnalyzer:
     
     def _generate_lip_summary(self, predictions: List[str]) -> Dict:
         """Generate lip engagement summary."""
-        lip_counts = Counter(predictions)
-        lip_counts_no_face = Counter({k: v for k, v in lip_counts.items() if k != "No Face"})
-        lip_total = sum(lip_counts_no_face.values())
-        
-        if lip_total > 0:
-            lip_summary = {
-                label: f"{lip_counts_no_face.get(label, 0)} frames ({(lip_counts_no_face.get(label, 0)/lip_total)*100:.1f}%)"
-                for label in ["Engaged", "Partially Engaged", "Not Engaged"]
-            }
-            lip_summary["majority_label"] = lip_counts_no_face.most_common(1)[0][0]
-        else:
-            lip_summary = {"No valid predictions": 0, "majority_label": "No Face"}
-        
-        return lip_summary
+        # This method is no longer used but kept for backward compatibility
+        return {}
     
     def _generate_micro_summary(self, eye_away: int, head_turn: int, total_frames: int) -> Dict:
         """Generate microexpression summary."""
@@ -340,13 +314,11 @@ interface = gr.Interface(
     inputs=gr.Video(label="Upload or record a 15s video (mp4/avi/mov)"),
     outputs=[
         gr.Textbox(label="Analysis Results (JSON)", lines=10),
-        gr.JSON(label="Lip Engagement Summary"),
         gr.JSON(label="Microexpression Summary")
     ],
     title="🎥 Advanced Video Analysis for SafeSpace",
     description="""
     Upload or record a video to analyze:
-    - **Engagement Level**: Based on lip movement and facial expressions
     - **Emotion Recognition**: Probability distribution across 7 emotions
     - **Stress Detection**: Weighted stress score based on emotional state
     - **Microexpressions**: Eye movement and head turn analysis