Update model.py

model.py

@@ -119,23 +119,14 @@ class ModelHandler:
         # --- END NEW ---
 
         # 4. Set Scheduler
-        # --- MODIFIED: Fine-tune LCMScheduler configuration ---
         print("Configuring fine-tuned LCMScheduler...")
-        
-        # Get the original config to use as a base
         scheduler_config = self.pipeline.scheduler.config
-        
-        # Apply our new, tuned settings
         scheduler_config['original_inference_steps'] = 75
         scheduler_config['timestep_spacing'] = "trailing"
         scheduler_config['timestep_scaling'] = 12.0
         scheduler_config['rescale_betas_zero_snr'] = True
-        
-        # Load the new scheduler from our modified config
         self.pipeline.scheduler = LCMScheduler.from_config(scheduler_config)
         print("  [OK] LCMScheduler fine-tuned.")
-        # --- END MODIFIED ---
-
 
         # 5. Load Adapters (IP-Adapter & LoRA)
         print("Loading Adapters (IP-Adapter & LoRA)...")
@@ -169,8 +160,9 @@ class ModelHandler:
         
         print("--- All models loaded successfully ---")
 
-    def get_face_embedding(self, image):
-        """Extracts face embedding, returns None if no face is found."""
+    # --- MODIFIED: Renamed function and changed return value ---
+    def get_face_info(self, image):
+        """Extracts the largest face, returns insightface result object."""
         if not self.face_analysis_loaded:
             return None
             
@@ -181,9 +173,11 @@ class ModelHandler:
             if len(faces) == 0:
                 return None
                 
+            # Sort by size (width * height) to find the main character
             faces = sorted(faces, key=lambda x: (x['bbox'][2]-x['bbox'][0])*(x['bbox'][3]-x['bbox'][1]), reverse=True)
             
-            return torch.tensor(faces[0].normed_embedding).unsqueeze(0)
+            # Return the largest face info
+            return faces[0]
         except Exception as e:
             print(f"Face embedding extraction failed: {e}")
             return None