All available YOLOv8-face models

deepface/detectors/FaceDetector.py

@@ -9,7 +9,7 @@ from deepface.detectors import (
     MtcnnWrapper,
     RetinaFaceWrapper,
     MediapipeWrapper,
-    Yolov8nfaceWrapper,
+    Yolov8faceWrapper,
 )
 
 
@@ -23,7 +23,9 @@ def build_model(detector_backend):
         "mtcnn": MtcnnWrapper.build_model,
         "retinaface": RetinaFaceWrapper.build_model,
         "mediapipe": MediapipeWrapper.build_model,
-        "yolov8n-face": Yolov8nfaceWrapper.build_model,
+        "yolov8-lite-t": Yolov8faceWrapper.build_model("yolov8-lite-t"),
+        "yolov8-lite-s": Yolov8faceWrapper.build_model("yolov8-lite-s"),
+        "yolov8n": Yolov8faceWrapper.build_model("yolov8n"),
     }
 
     if not "face_detector_obj" in globals():
@@ -63,7 +65,9 @@ def detect_faces(face_detector, detector_backend, img, align=True):
         "mtcnn": MtcnnWrapper.detect_face,
         "retinaface": RetinaFaceWrapper.detect_face,
         "mediapipe": MediapipeWrapper.detect_face,
-        "yolov8n-face": Yolov8nfaceWrapper.detect_face,
+        "yolov8-lite-t": Yolov8faceWrapper.detect_face,
+        "yolov8-lite-s": Yolov8faceWrapper.detect_face,
+        "yolov8n": Yolov8faceWrapper.detect_face,
     }
 
     detect_face_fn = backends.get(detector_backend)

deepface/detectors/Yolov8faceWrapper.py

@@ -0,0 +1,64 @@
+from deepface.detectors import FaceDetector
+
+PATHS = {
+    "yolov8-lite-t": "/.deepface/weights/yolov8-lite-t.pt",
+    "yolov8-lite-s": "/.deepface/weights/yolov8-lite-s.pt",
+    "yolov8n": "/.deepface/weights/yolov8n-face.pt",
+}
+
+BASE_URL = "https://drive.google.com/uc?id="
+
+IDS = {
+    "yolov8-lite-t": "1vFMGW8xtRVo9bfC9yJVWWGY7vVxbLh94",
+    "yolov8-lite-s": "1ckpBT8KfwURTvTm5pa-cMC89A0V5jbaq",
+    "yolov8n": "1qcr9DbgsX3ryrz2uU8w4Xm3cOrRywXqb",
+}
+
+
+def build_model(model: str):
+    """Function factory for YOLO models"""
+    from deepface.commons.functions import get_deepface_home
+
+    func_weights_path = f"{get_deepface_home()}{PATHS[model]}"
+    func_url = f"{BASE_URL}{IDS[model]}"
+
+    def _build_model(weights_path: str = func_weights_path, url: str = func_url):
+        import gdown
+        import os
+
+        from ultralytics import YOLO
+
+        if not os.path.isfile(weights_path):
+            gdown.download(url, weights_path, quiet=False)
+            print(f"Downloaded YOLO model {os.path.basename(PATHS[model])}")
+
+        # return face_detector
+        return YOLO(weights_path)
+
+    return _build_model
+
+
+def detect_face(face_detector, img, align=False):
+    resp = []
+
+    results = face_detector.predict(img, verbose=False, show=True, conf=0.25)[0]
+
+    for result in results:
+        x, y, w, h = result.boxes.xywh.tolist()[0]
+        confidence = result.boxes.conf.tolist()[0]
+
+        x, y, w, h = int(x - w / 2), int(y - h / 2), int(w), int(h)
+        detected_face = img[y : y + h, x : x + w].copy()
+
+        if align:
+            # Extract landmarks
+            left_eye, right_eye, _, _, _ = result.keypoints.tolist()
+            # Check the landmarks confidence before alignment
+            if left_eye[2] > 0.5 and right_eye[2] > 0.5:
+                detected_face = FaceDetector.alignment_procedure(
+                    detected_face, left_eye[:2], right_eye[:2]
+                )
+
+        resp.append((detected_face, [x, y, w, h], confidence))
+
+    return resp

deepface/detectors/Yolov8nfaceWrapper.py

@@ -1,46 +0,0 @@
-from deepface.detectors import FaceDetector
-
-
-def build_model():
-    import gdown
-    import os
-
-    from ultralytics import YOLO
-
-    from deepface.commons.functions import get_deepface_home
-
-    weights_path = f"{get_deepface_home()}/.deepface/weights/yolov8n-face.pt"
-
-    if not os.path.isfile(weights_path):
-        url = "https://drive.google.com/uc?id=1qcr9DbgsX3ryrz2uU8w4Xm3cOrRywXqb"
-        gdown.download(url, weights_path, quiet=False)
-        print("Downloaded YOLO model yolo8vn-face.pt")
-
-    # return face_detector
-    return YOLO(weights_path)
-
-
-def detect_face(face_detector, img, align=False):
-    resp = []
-
-    results = face_detector.predict(img, verbose=False, show=True, conf=0.25)[0]
-
-    for result in results:
-        x, y, w, h = result.boxes.xywh.tolist()[0]
-        confidence = result.boxes.conf.tolist()[0]
-
-        x, y, w, h = int(x - w / 2), int(y - h / 2), int(w), int(h)
-        detected_face = img[y : y + h, x : x + w].copy()
-
-        if align:
-            # Extract landmarks
-            left_eye, right_eye, _, _, _ = result.keypoints.tolist()
-            # Check the landmarks confidence before alignment
-            if left_eye[2] > 0.5 and right_eye[2] > 0.5:
-                detected_face = FaceDetector.alignment_procedure(
-                    detected_face, left_eye[:2], right_eye[:2]
-                )
-
-        resp.append((detected_face, [x, y, w, h], confidence))
-
-    return resp