Merge pull request serengil#758 from Vincent-Stragier/YOLOv8

Integration of YOLOv8-face close serengil#732

README.md

@@ -194,7 +194,7 @@ Age model got ± 4.65 MAE; gender model got 97.44% accuracy, 96.29% precision an
 
 **Face Detectors** - [`Demo`](https://youtu.be/GZ2p2hj2H5k)
 
-Face detection and alignment are important early stages of a modern face recognition pipeline. Experiments show that just alignment increases the face recognition accuracy almost 1%. [`OpenCV`](https://sefiks.com/2020/02/23/face-alignment-for-face-recognition-in-python-within-opencv/), [`SSD`](https://sefiks.com/2020/08/25/deep-face-detection-with-opencv-in-python/), [`Dlib`](https://sefiks.com/2020/07/11/face-recognition-with-dlib-in-python/),  [`MTCNN`](https://sefiks.com/2020/09/09/deep-face-detection-with-mtcnn-in-python/), [`RetinaFace`](https://sefiks.com/2021/04/27/deep-face-detection-with-retinaface-in-python/) and [`MediaPipe`](https://sefiks.com/2022/01/14/deep-face-detection-with-mediapipe/) detectors are wrapped in deepface.
+Face detection and alignment are important early stages of a modern face recognition pipeline. Experiments show that just alignment increases the face recognition accuracy almost 1%. [`OpenCV`](https://sefiks.com/2020/02/23/face-alignment-for-face-recognition-in-python-within-opencv/), [`SSD`](https://sefiks.com/2020/08/25/deep-face-detection-with-opencv-in-python/), [`Dlib`](https://sefiks.com/2020/07/11/face-recognition-with-dlib-in-python/),  [`MTCNN`](https://sefiks.com/2020/09/09/deep-face-detection-with-mtcnn-in-python/), [`RetinaFace`](https://sefiks.com/2021/04/27/deep-face-detection-with-retinaface-in-python/), [`MediaPipe`](https://sefiks.com/2022/01/14/deep-face-detection-with-mediapipe/) and [`YOLOv8 Face`](https://github.com/derronqi/yolov8-face) detectors are wrapped in deepface.
 
 <p align="center"><img src="https://raw.githubusercontent.com/serengil/deepface/master/icon/detector-portfolio-v3.jpg" width="95%" height="95%"></p>
 
@@ -207,7 +207,8 @@ backends = [
   'dlib', 
   'mtcnn', 
   'retinaface', 
-  'mediapipe'
+  'mediapipe',
+  'yolov8',
 ]
 
 #face verification

deepface/DeepFace.py

@@ -116,7 +116,7 @@ def verify(
             This might be convenient for low resolution images.
 
             detector_backend (string): set face detector backend to opencv, retinaface, mtcnn, ssd,
-            dlib or mediapipe
+            dlib, mediapipe or yolov8.
 
             align (boolean): alignment according to the eye positions.
 
@@ -251,7 +251,7 @@ def analyze(
             resolution images.
 
             detector_backend (string): set face detector backend to opencv, retinaface, mtcnn, ssd,
-            dlib or mediapipe.
+            dlib, mediapipe or yolov8.
 
             align (boolean): alignment according to the eye positions.
 
@@ -429,7 +429,7 @@ def find(
             resolution images.
 
             detector_backend (string): set face detector backend to opencv, retinaface, mtcnn, ssd,
-            dlib or mediapipe
+            dlib, mediapipe or yolov8.
 
             align (boolean): alignment according to the eye positions.
 
@@ -456,6 +456,7 @@ def find(
     file_name = file_name.replace("-", "_").lower()
 
     if path.exists(db_path + "/" + file_name):
+
         if not silent:
             print(
                 f"WARNING: Representations for images in {db_path} folder were previously stored"
@@ -640,7 +641,7 @@ def represent(
             This might be convenient for low resolution images.
 
             detector_backend (string): set face detector backend to opencv, retinaface, mtcnn, ssd,
-            dlib or mediapipe
+            dlib, mediapipe or yolov8.
 
             align (boolean): alignment according to the eye positions.
 
@@ -725,7 +726,7 @@ def stream(
             model_name (string): VGG-Face, Facenet, Facenet512, OpenFace, DeepFace, DeepID, Dlib,
             ArcFace, SFace
 
-            detector_backend (string): opencv, retinaface, mtcnn, ssd, dlib or mediapipe
+            detector_backend (string): opencv, retinaface, mtcnn, ssd, dlib, mediapipe or yolov8.
 
             distance_metric (string): cosine, euclidean, euclidean_l2
 

deepface/detectors/FaceDetector.py

@@ -9,11 +9,11 @@
     MtcnnWrapper,
     RetinaFaceWrapper,
     MediapipeWrapper,
+    YoloWrapper,
 )
 
 
 def build_model(detector_backend):
-
     global face_detector_obj  # singleton design pattern
 
     backends = {
@@ -23,6 +23,7 @@ def build_model(detector_backend):
         "mtcnn": MtcnnWrapper.build_model,
         "retinaface": RetinaFaceWrapper.build_model,
         "mediapipe": MediapipeWrapper.build_model,
+        "yolov8": YoloWrapper.build_model,
     }
 
     if not "face_detector_obj" in globals():
@@ -42,27 +43,30 @@ def build_model(detector_backend):
 
 
 def detect_face(face_detector, detector_backend, img, align=True):
-
     obj = detect_faces(face_detector, detector_backend, img, align)
 
     if len(obj) > 0:
         face, region, confidence = obj[0]  # discard multiple faces
+
+    # If no face is detected, set face to None,
+    # image region to full image, and confidence to 0.
     else:  # len(obj) == 0
         face = None
         region = [0, 0, img.shape[1], img.shape[0]]
+        confidence = 0
 
     return face, region, confidence
 
 
 def detect_faces(face_detector, detector_backend, img, align=True):
-
     backends = {
         "opencv": OpenCvWrapper.detect_face,
         "ssd": SsdWrapper.detect_face,
         "dlib": DlibWrapper.detect_face,
         "mtcnn": MtcnnWrapper.detect_face,
         "retinaface": RetinaFaceWrapper.detect_face,
         "mediapipe": MediapipeWrapper.detect_face,
+        "yolov8": YoloWrapper.detect_face,
     }
 
     detect_face_fn = backends.get(detector_backend)
@@ -76,7 +80,6 @@ def detect_faces(face_detector, detector_backend, img, align=True):
 
 
 def alignment_procedure(img, left_eye, right_eye):
-
     # this function aligns given face in img based on left and right eye coordinates
 
     left_eye_x, left_eye_y = left_eye
@@ -104,7 +107,6 @@ def alignment_procedure(img, left_eye, right_eye):
     # apply cosine rule
 
     if b != 0 and c != 0:  # this multiplication causes division by zero in cos_a calculation
-
         cos_a = (b * b + c * c - a * a) / (2 * b * c)
         angle = np.arccos(cos_a)  # angle in radian
         angle = (angle * 180) / math.pi  # radian to degree

deepface/detectors/YoloWrapper.py

@@ -0,0 +1,62 @@
+from deepface.detectors import FaceDetector
+
+# Model's weights paths
+PATH = "/.deepface/weights/yolov8n-face.pt"
+
+# Google Drive URL
+WEIGHT_URL = "https://drive.google.com/uc?id=1qcr9DbgsX3ryrz2uU8w4Xm3cOrRywXqb"
+
+# Confidence thresholds for landmarks detection
+# used in alignment_procedure function
+LANDMARKS_CONFIDENCE_THRESHOLD = 0.5
+
+
+def build_model():
+    """Build YOLO (yolov8n-face) model"""
+    import gdown
+    import os
+
+    # Import the Ultralytics YOLO model
+    from ultralytics import YOLO
+
+    from deepface.commons.functions import get_deepface_home
+    weight_path = f"{get_deepface_home()}{PATH}"
+
+    # Download the model's weights if they don't exist
+    if not os.path.isfile(weight_path):
+        gdown.download(WEIGHT_URL, weight_path, quiet=False)
+        print(f"Downloaded YOLO model {os.path.basename(weight_path)}")
+
+    # Return face_detector
+    return YOLO(weight_path)
+
+
+def detect_face(face_detector, img, align=False):
+    resp = []
+
+    # Detect faces
+    results = face_detector.predict(
+        img, verbose=False, show=False, conf=0.25)[0]
+
+    # For each face, extract the bounding box, the landmarks and confidence
+    for result in results:
+        # Extract the bounding box and the confidence
+        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] > LANDMARKS_CONFIDENCE_THRESHOLD and
+                    right_eye[2] > LANDMARKS_CONFIDENCE_THRESHOLD):
+                detected_face = FaceDetector.alignment_procedure(
+                    detected_face, left_eye[:2], right_eye[:2]
+                )
+
+        resp.append((detected_face, [x, y, w, h], confidence))
+
+    return resp

requirements_additional.txt

@@ -1,3 +1,4 @@
 opencv-contrib-python>=4.3.0.36
 mediapipe>=0.8.7.3
-dlib>=19.20.0
+dlib>=19.20.0
+ultralytics @ git+https://github.com/derronqi/yolov8-face.git@b623989575bdb78601b5ca717851e3d63ca9e01c