Custom object detection dataset in Computer Vision - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to load images from a folder using OpenCV.

Computer Vision

import cv2
image = cv2.imread([1])
print(image.shape)

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

A'image.jpg'

Bimage.jpg

C['image.jpg']

Dimage

Attempts:

3 left

💡 Hint

Common Mistakes

Not putting the filename in quotes causes an error.

Passing a list instead of a string.

You need to provide the image filename as a string to cv2.imread(). So, it must be in quotes.

2fill in blank

medium

Complete the code to create a list of image file paths from a directory using pathlib.

Computer Vision

from pathlib import Path
image_dir = Path('images')
image_files = list(image_dir.glob([1]))
print(len(image_files))

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

A'*.jpg'

B'*jpg'

C'jpg*'

D'*.jpeg'

Attempts:

3 left

💡 Hint

Common Mistakes

Missing the * wildcard causes no files to be found.

Using incorrect file extension pattern.

The glob method needs a pattern string with a wildcard and file extension, like '*.jpg', to find all JPG files.

3fill in blank

hard

Fix the error in this code that reads bounding box annotations from a CSV file.

Computer Vision

import pandas as pd
annotations = pd.read_csv([1])
print(annotations.head())

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

Acsv.annotations

Bannotations.csv

Cannotations

D'annotations.csv'

Attempts:

3 left

💡 Hint

Common Mistakes

Passing a variable name instead of a string filename.

Forgetting quotes around the filename.

The filename must be a string in quotes when passed to pd.read_csv().

4fill in blank

hard

Fill both blanks to create a dictionary of image filenames and their bounding boxes from a DataFrame.

Computer Vision

bbox_dict = {row[[1]]: (row['xmin'], row['ymin'], row['xmax'], row[[2]]) for _, row in annotations.iterrows()}

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

A'filename'

B'xmax'

D'ymax'

Attempts:

3 left

💡 Hint

Common Mistakes

Using wrong column names for keys or bounding box coordinates.

Mixing up xmax and ymax columns.

The dictionary key is the filename from the 'filename' column. The bounding box tuple uses 'xmin', 'ymin', 'xmax', and 'ymax' columns. Here, the last blank should be 'ymax'.

5fill in blank

hard

Fill all three blanks to filter annotations for a specific class and create a list of bounding boxes.

Computer Vision

class_name = 'cat'
filtered = annotations[annotations[[1]] == [2]]
bboxes = [(row['xmin'], row['ymin'], row[[3]], row['ymax']) for _, row in filtered.iterrows()]

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

A'class'

B'cat'

C'xmax'

D'filename'

Attempts:

3 left

💡 Hint

Common Mistakes

Filtering by wrong column name.

Using wrong bounding box coordinate names.

We filter rows where the 'class' column equals 'cat'. Then we extract bounding boxes using 'xmin', 'ymin', 'xmax', and 'ymax'. The third blank is 'xmax'.