diff --git a/models/Model_Quick96/Model.py b/models/Model_Quick96/Model.py
index dba5738..8b999af 100644
--- a/models/Model_Quick96/Model.py
+++ b/models/Model_Quick96/Model.py
@@ -307,7 +307,12 @@ class QModel(ModelBase):
 
         st_m = []
         for i in range(n_samples):
-            ar = S[i]*target_srcm[i], SS[i], D[i]*target_dstm[i], DD[i]*DDM[i], SD[i]*(DDM[i]*SDM[i])
+            SM = S[i]*target_srcm[i]
+            DM = D[i]*target_dstm[i]
+            if filenames is not None and len(filenames) > 0:
+                SM = label_face_filename(SM, filenames[0][i])
+                DM = label_face_filename(DM, filenames[0][i])
+            ar = SM, SS[i], DM, DD[i]*DDM[i], SD[i]*(DDM[i]*SDM[i])
             st_m.append ( np.concatenate ( ar, axis=1) )
 
         result += [ ('Quick96 masked', np.concatenate (st_m, axis=0 )), ]
diff --git a/models/Model_SAEHD/Model.py b/models/Model_SAEHD/Model.py
index 5b87442..2fb3a75 100644
--- a/models/Model_SAEHD/Model.py
+++ b/models/Model_SAEHD/Model.py
@@ -983,8 +983,12 @@ class SAEHDModel(ModelBase):
             st_m = []
             for i in range(n_samples):
                 SD_mask = DDM[i]*SDM[i] if self.face_type < FaceType.HEAD else SDM[i]
-
-                ar = S[i]*target_srcm[i], SS[i]*SSM[i], D[i]*target_dstm[i], DD[i]*DDM[i], SD[i]*SD_mask
+                SM = S[i]*target_srcm[i]
+                DM = D[i]*target_dstm[i]
+                if filenames is not None and len(filenames) > 0:
+                    SM = label_face_filename(SM, filenames[0][i])
+                    DM = label_face_filename(DM, filenames[0][i])
+                ar = SM, SS[i]*SSM[i], DM, DD[i]*DDM[i], SD[i]*SD_mask
                 st_m.append ( np.concatenate ( ar, axis=1) )
 
             result += [ ('SAEHD masked', np.concatenate (st_m, axis=0 )), ]
diff --git a/models/Model_XSeg/Model.py b/models/Model_XSeg/Model.py
index 4f536cf..825e25c 100644
--- a/models/Model_XSeg/Model.py
+++ b/models/Model_XSeg/Model.py
@@ -198,7 +198,7 @@ class XSegModel(ModelBase):
         return ( ('loss', np.mean(loss) ), )
 
     #override
-    def onGetPreview(self, samples, for_history=False):
+    def onGetPreview(self, samples, for_history=False, filenames=None):
         n_samples = min(4, self.get_batch_size(), 800 // self.resolution )
         
         if self.pretrain: