FaceMesh

app.py

@@ -28,7 +28,7 @@ faces_cache = {"boxes": None, "text": "Initializing...", "counter": 0}
 # -----------------------------
 # Initialize Models and Helpers
 # -----------------------------
-# MediaPipe Pose and Face Detection
+# MediaPipe Pose, Face Detection, and Face Mesh
 mp_pose = mp.solutions.pose
 pose = mp_pose.Pose()
 mp_drawing = mp.solutions.drawing_utils
@@ -51,6 +51,7 @@ object_categories = FasterRCNN_ResNet50_FPN_Weights.DEFAULT.meta["categories"]
 
 # -----------------------------
 # Facial Recognition Model (Marltgap/FaceTransformerOctupletLoss ONNX)
+# (No longer used in the UI; kept here for reference)
 # -----------------------------
 facial_recognition_onnx = rt.InferenceSession("FaceTransformerOctupletLoss.onnx", providers=rt.get_available_providers())
 
@@ -149,6 +150,39 @@ def compute_faces_overlay(image):
         text = "No faces detected"
     return boxes, text
 
+# -----------------------------
+# New Facemesh Functions
+# -----------------------------
+def compute_facemesh_overlay(image):
+    """
+    Uses MediaPipe Face Mesh to detect and draw facial landmarks.
+    """
+    frame_bgr = cv2.cvtColor(np.array(image), cv2.COLOR_RGB2BGR)
+    h, w, _ = frame_bgr.shape
+    # Initialize Face Mesh in static mode
+    face_mesh = mp.solutions.face_mesh.FaceMesh(
+        static_image_mode=True, max_num_faces=1, refine_landmarks=True, min_detection_confidence=0.5
+    )
+    results = face_mesh.process(cv2.cvtColor(frame_bgr, cv2.COLOR_BGR2RGB))
+    if results.multi_face_landmarks:
+        for face_landmarks in results.multi_face_landmarks:
+            for landmark in face_landmarks.landmark:
+                x = int(landmark.x * w)
+                y = int(landmark.y * h)
+                cv2.circle(frame_bgr, (x, y), 1, (0, 255, 0), -1)
+        text = "Facemesh detected"
+    else:
+        text = "No facemesh detected"
+    face_mesh.close()
+    return frame_bgr, text
+
+def analyze_facemesh(image):
+    annotated_image, text = compute_facemesh_overlay(image)
+    return annotated_image, f"<div style='color: lime !important;'>Facemesh Analysis: {text}</div>"
+
+# -----------------------------
+# (Retained) Facial Recognition Function (Not used in UI anymore)
+# -----------------------------
 def compute_facial_recognition_vector(image):
     """
     Detects a face using MediaPipe, crops and resizes it to 112x112, then computes its embedding
@@ -232,6 +266,7 @@ def analyze_faces_current(image):
         output = draw_boxes_overlay(output, faces_cache["boxes"], (0, 0, 255))
     return output, f"<div style='color: lime !important;'>Face Detection: {faces_cache['text']}</div>"
 
+# (The old facial recognition analysis function is retained below but not linked to any UI tab)
 def analyze_facial_recognition(image):
     # Compute and return the facial vector (and the cropped face)
     face_crop, vector_str = compute_facial_recognition_vector(image)
@@ -333,12 +368,15 @@ faces_interface = gr.Interface(
     live=False
 )
 
-facial_recognition_interface = gr.Interface(
-    fn=analyze_facial_recognition,
-    inputs=gr.Image(label="Upload a Face Image for Facial Recognition"),
-    outputs=[gr.Image(type="numpy", label="Cropped Face"), gr.HTML(label="Facial Recognition")],
-    title="Facial Recognition",
-    description="Extracts and outputs the facial vector using the Marltgap FaceTransformerOctupletLoss ONNX model.",
+# -----------------------------
+# New Facemesh Interface (Replaces the old Facial Recognition tab)
+# -----------------------------
+facemesh_interface = gr.Interface(
+    fn=analyze_facemesh,
+    inputs=gr.Image(label="Upload an Image for Facemesh"),
+    outputs=[gr.Image(type="numpy", label="Annotated Output"), gr.HTML(label="Facemesh Analysis")],
+    title="Facemesh",
+    description="Detects facial landmarks using MediaPipe Face Mesh.",
     live=False
 )
 
@@ -357,7 +395,7 @@ tabbed_interface = gr.TabbedInterface(
         emotion_interface,
         objects_interface,
         faces_interface,
-        facial_recognition_interface,
+        facemesh_interface,
         all_interface
     ],
     tab_names=[
@@ -365,7 +403,7 @@ tabbed_interface = gr.TabbedInterface(
         "Emotion",
         "Objects",
         "Faces",
-        "Facial Recognition",
+        "Facemesh",
         "All Inferences"
     ]
 )
@@ -376,7 +414,7 @@ tabbed_interface = gr.TabbedInterface(
 demo = gr.Blocks(css=custom_css)
 with demo:
     gr.Markdown("<h1 class='gradio-title' style='color: #32CD32;'>Multi-Analysis Image App</h1>")
-    gr.Markdown("<p class='gradio-description' style='color: #32CD32;'>Upload an image to run high-tech analysis for posture, emotions, objects, faces, and facial embeddings.</p>")
+    gr.Markdown("<p class='gradio-description' style='color: #32CD32;'>Upload an image to run high-tech analysis for posture, emotions, objects, faces, and facemesh landmarks.</p>")
     tabbed_interface.render()
 
 if __name__ == "__main__":