Face Detection

1. Experiment Objective

Use MediaPipe Face Mesh to detect facial landmarks in the camera image, generating 468 facial mesh points and drawing contour connections.

2. Experiment Principle

1. Key Terms

2. Technical Architecture

Terminal 1: start agent + camera + joint models
Terminal 2: start face detection

3. Experiment Steps

1. Hardware Preparation

2. Start the agent + camera + joint models (Terminal 1)

xxxxxxxxxx
ros2 launch microros_v2_bringup car_base.launch.py

3. Start face detection (Terminal 2)

xxxxxxxxxx
ros2 launch microros_v2_mediapipe_pkg face_mesh.launch.py

4. Operating Instructions

Press q to exit.

4. Observed Results

• The face is covered by 468 white dots, with triangular faces forming a fine facial surface mesh
• Facial contour lines (eyebrows, eyes, lips, outer face contour) are annotated in distinctive colors

5. Code Walkthrough

Source code paths:

• Launch: ~/microros_ws/src/microros_v2_mediapipe_pkg/launch/face_mesh.launch.py
• Node: ~/microros_ws/src/microros_v2_mediapipe_pkg/microros_v2_mediapipe_pkg/FaceMesh.py

1. Core Node Analysis

​x
# Source file: ~/microros_ws/src/microros_v2_mediapipe_pkg/microros_v2_mediapipe_pkg/FaceMesh.py
class FaceMesh(Node):
    def __init__(self):
        super().__init__("face_mesh_node")
        self.mp_face_mesh = mp.solutions.face_mesh
        self.face_mesh = self.mp_face_mesh.FaceMesh(
            static_image_mode=False,
            max_num_faces=1,
            min_detection_confidence=0.5,
            min_tracking_confidence=0.5)
​
    def image_callback(self, msg):
        frame = self.bridge.imgmsg_to_cv2(msg, 'bgr8')
        results = self.face_mesh.process(cv.cvtColor(frame, cv.COLOR_BGR2RGB))
        if results.multi_face_landmarks:
            for face_landmarks in results.multi_face_landmarks:
                self.mp_draw.draw_landmarks(
                    frame, face_landmarks,
                    self.mp_face_mesh.FACEMESH_TESSELATION, ...)
        cv.imshow(self.window_name, frame)

Key logic:

• mp.solutions.face_mesh.FaceMesh(): 468-point facial mesh detector
• max_num_faces=1: detects 1 face by default; increase it to support multiple faces
• FACEMESH_TESSELATION: triangular face topology that represents the facial surface more finely than skeleton connections
• Supports FACEMESH_CONTOURS (contour mode), which only draws facial contour lines (eyebrows, eyes, lips, outer contour) to reduce visual complexity

2. Key Parameters

Parameter definition files:

• Launch: ~/microros_ws/src/microros_v2_mediapipe_pkg/launch/face_mesh.launch.py
• Node: ~/microros_ws/src/microros_v2_mediapipe_pkg/microros_v2_mediapipe_pkg/FaceMesh.py

3. Node Description

6. Frequently Asked Questions