diff --git a/roop/core.py b/roop/core.py
index 7e5a46fbd..71c29f752 100755
--- a/roop/core.py
+++ b/roop/core.py
@@ -38,6 +38,7 @@ def parse_args() -> None:
     program.add_argument('--skip-audio', help='skip target audio', dest='skip_audio', action='store_true')
     program.add_argument('--many-faces', help='process every face', dest='many_faces', action='store_true')
     program.add_argument('--reference-face-position', help='position of the reference face', dest='reference_face_position', type=int, default=0)
+    program.add_argument('--reference-face-path', help='reference face path', dest='face_reference_path')
     program.add_argument('--reference-frame-number', help='number of the reference frame', dest='reference_frame_number', type=int, default=0)
     program.add_argument('--similar-face-distance', help='face distance used for recognition', dest='similar_face_distance', type=float, default=0.85)
     program.add_argument('--temp-frame-format', help='image format used for frame extraction', dest='temp_frame_format', default='png', choices=['jpg', 'png'])
@@ -62,6 +63,7 @@ def parse_args() -> None:
     roop.globals.many_faces = args.many_faces
     roop.globals.reference_face_position = args.reference_face_position
     roop.globals.reference_frame_number = args.reference_frame_number
+    roop.globals.reference_face_path = args.face_reference_path
     roop.globals.similar_face_distance = args.similar_face_distance
     roop.globals.temp_frame_format = args.temp_frame_format
     roop.globals.temp_frame_quality = args.temp_frame_quality
@@ -71,7 +73,6 @@ def parse_args() -> None:
     roop.globals.execution_providers = decode_execution_providers(args.execution_provider)
     roop.globals.execution_threads = args.execution_threads
 
-
 def encode_execution_providers(execution_providers: List[str]) -> List[str]:
     return [execution_provider.replace('ExecutionProvider', '').lower() for execution_provider in execution_providers]
 
diff --git a/roop/globals.py b/roop/globals.py
index 3eca8d0d0..2f614e214 100644
--- a/roop/globals.py
+++ b/roop/globals.py
@@ -11,6 +11,7 @@
 many_faces: Optional[bool] = None
 reference_face_position: Optional[int] = None
 reference_frame_number: Optional[int] = None
+reference_face_path: Optional[str] = None
 similar_face_distance: Optional[float] = None
 temp_frame_format: Optional[str] = None
 temp_frame_quality: Optional[int] = None
diff --git a/roop/processors/frame/face_swapper.py b/roop/processors/frame/face_swapper.py
index da68956ef..02d5d2036 100644
--- a/roop/processors/frame/face_swapper.py
+++ b/roop/processors/frame/face_swapper.py
@@ -93,8 +93,19 @@ def process_image(source_path: str, target_path: str, output_path: str) -> None:
 
 
 def process_video(source_path: str, temp_frame_paths: List[str]) -> None:
+
+    if roop.globals.reference_face_path:
+        reference_face = get_one_face(cv2.imread(roop.globals.reference_face_path))
+        set_face_reference(reference_face)
+
+
     if not roop.globals.many_faces and not get_face_reference():
-        reference_frame = cv2.imread(temp_frame_paths[roop.globals.reference_frame_number])
-        reference_face = get_one_face(reference_frame, roop.globals.reference_face_position)
+        # recorro los temp_frame_paths hasta que reference_face sea distinto de nulo.
+        for temp_frame_path in temp_frame_paths:
+            temp_frame = cv2.imread(temp_frame_path)
+            reference_face = get_one_face(temp_frame, roop.globals.reference_face_position)
+            if reference_face:
+                break
+
         set_face_reference(reference_face)
     roop.processors.frame.core.process_video(source_path, temp_frame_paths, process_frames)