Inpainting and outpainting in Prompt Engineering / GenAI - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to load an image and prepare a mask for inpainting.

Prompt Engineering / GenAI

from PIL import Image
image = Image.open('input.jpg')
mask = Image.new('L', image.size, [1])

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

A0

B255

C128

D1

Attempts:

3 left

💡 Hint

Common Mistakes

Using 255 instead of 0 for the initial mask value.

Confusing mask color values.

The mask for inpainting is usually black (0) where no change is needed and white (255) where inpainting is applied. Here, we start with a black mask.

2fill in blank

medium

Complete the code to apply inpainting using OpenCV.

Prompt Engineering / GenAI

import cv2
image = cv2.imread('input.jpg')
mask = cv2.imread('mask.png', 0)
inpainted = cv2.inpaint(image, mask, [1], cv2.INPAINT_TELEA)

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

A10

B0

C3

D1

Attempts:

3 left

💡 Hint

Common Mistakes

Using 0 radius which disables inpainting.

Using too large radius causing blurring.

The radius parameter controls the neighborhood size for inpainting. A small radius like 3 is typical.

3fill in blank

hard

Fix the error in the code to perform outpainting by extending the image canvas.

Prompt Engineering / GenAI

import numpy as np
from PIL import Image
image = Image.open('input.jpg')
width, height = image.size
new_width = width + 100
new_height = height + 50
new_image = Image.new('RGB', (new_width, new_height), [1])
new_image.paste(image, (50, 25))

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

A(255, 0, 0)

B(0, 0, 0)

C(128, 128, 128)

D(255, 255, 255)

Attempts:

3 left

💡 Hint

Common Mistakes

Using black background causing harsh edges.

Using a single integer instead of a tuple for color.

The new canvas background color is white (255,255,255) to blend well for outpainting.

4fill in blank

hard

Fill both blanks to create a mask dictionary for inpainting and outpainting regions.

Prompt Engineering / GenAI

masks = {
    'inpainting': [1],
    'outpainting': [2]
}

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

A{'region': 'center', 'mask': mask_inpaint}

B{'region': 'border', 'mask': mask_outpaint}

C{'region': 'edges', 'mask': mask_outpaint}

D{'region': 'center', 'mask': mask_outpaint}

Attempts:

3 left

💡 Hint

Common Mistakes

Swapping inpainting and outpainting masks.

Using the same mask for both regions.

Inpainting mask targets the center region, outpainting mask targets the border region.

5fill in blank

hard

Fill all three blanks to define a function that applies inpainting or outpainting based on mode.

Prompt Engineering / GenAI

def apply_painting(image, mask, mode):
    if mode == 'inpainting':
        method = [1]
    elif mode == 'outpainting':
        method = [2]
    else:
        raise ValueError('Invalid mode')
    return cv2.inpaint(image, mask, 3, method)  # radius fixed at 3

result = apply_painting(img, mask, [3])

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

Acv2.INPAINT_TELEA

Bcv2.INPAINT_NS

C'inpainting'

D'outpainting'

Attempts:

3 left

💡 Hint

Common Mistakes

Using the same method for both modes.

Passing mode as a variable instead of string.

Use TELEA method for inpainting, NS method for outpainting, and call function with mode 'outpainting'.