Hand and face landmark detection in Computer Vision - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to import the MediaPipe Hands solution.

Computer Vision

import mediapipe as mp
mp_hands = mp.solutions.[1]

Drag options to blanks, or click blank then click option'

Ahands

Bface_mesh

Cdrawing_utils

Dpose

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'face_mesh' instead of 'hands' for hand detection.

Using 'pose' which is for body pose estimation.

Using 'drawing_utils' which is for drawing landmarks, not detection.

The MediaPipe Hands solution is accessed via mp.solutions.hands. This module provides hand landmark detection.

2fill in blank

medium

Complete the code to initialize the Hands model with static image mode enabled.

Computer Vision

with mp_hands.Hands(static_image_mode=[1], max_num_hands=2) as hands:

Drag options to blanks, or click blank then click option'

AFalse

BNone

CTrue

D0

Attempts:

3 left

💡 Hint

Common Mistakes

Setting static_image_mode to False when processing single images.

Using None or 0 which are invalid for this parameter.

Setting static_image_mode=True tells the model to treat input as static images, which is important for processing single images instead of video streams.

3fill in blank

hard

Fix the error in accessing the hand landmarks from the results object.

Computer Vision

if results.[1]:
    for hand_landmarks in results.multi_hand_landmarks:
        print(hand_landmarks)

Drag options to blanks, or click blank then click option'

Amulti_hand_landmarks

Bhand_landmarks

Clandmarks

Dmulti_landmarks

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'hand_landmarks' which is not an attribute of results.

Using 'landmarks' or 'multi_landmarks' which do not exist.

The detected hand landmarks are stored in results.multi_hand_landmarks. This attribute contains a list of landmarks for each detected hand.

4fill in blank

hard

Fill both blanks to draw hand landmarks on the image using MediaPipe drawing utilities.

Computer Vision

mp_drawing.[1](image, hand_landmarks, mp_hands.[2])

Drag options to blanks, or click blank then click option'

Adraw_landmarks

Bdraw_detection

CHAND_CONNECTIONS

Dface_mesh

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'draw_detection' which is for object detection, not landmarks.

Using 'face_mesh' which is unrelated to hands.

The function draw_landmarks is used to draw landmarks on images. The connections for hands are accessed via mp_hands.HAND_CONNECTIONS.

5fill in blank

hard

Fill all three blanks to create a dictionary of landmark coordinates normalized to image size.

Computer Vision

landmark_dict = [1]: {'x': [2].x * image_width, 'y': [2].y * image_height} for [1], [2] in [3](hand_landmarks.landmark)

Drag options to blanks, or click blank then click option'

Ai

Blandmark_dict

Cenumerate

Didx

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'landmark_dict' as a variable inside the comprehension.

Mixing variable names inconsistently.

Not using 'enumerate' to get index and value.

This dictionary comprehension uses enumerate to get the index idx and landmark i. The index is used as the key, and the landmark coordinates are scaled by image width and height.