Depth estimation basics in Computer Vision - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to load an image for depth estimation.

Computer Vision

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

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

Adepth_map.png

B'depth_map'

C'depth_map.png'

Ddepth_map

Attempts:

3 left

💡 Hint

Common Mistakes

Forgetting quotes around the filename.

Passing a variable name without defining it.

The cv2.imread function requires the filename as a string with quotes.

2fill in blank

medium

Complete the code to convert the image to grayscale for depth processing.

Computer Vision

gray_image = cv2.cvtColor(image, [1])
print(gray_image.shape)

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

Acv2.COLOR_BGR2GRAY

Bcv2.COLOR_BGR2RGB

Ccv2.COLOR_RGB2GRAY

Dcv2.COLOR_GRAY2BGR

Attempts:

3 left

💡 Hint

Common Mistakes

Using RGB conversion instead of grayscale.

Using grayscale to BGR conversion by mistake.

To convert a color image to grayscale, use cv2.COLOR_BGR2GRAY.

3fill in blank

hard

Fix the error in the code to normalize the depth map between 0 and 1.

Computer Vision

depth_normalized = (depth_map - depth_map.min()) / [1]
print(depth_normalized.min(), depth_normalized.max())

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

Adepth_map.min()

Bdepth_map.max() - depth_map.min()

Cdepth_map.max()

Ddepth_map.mean()

Attempts:

3 left

💡 Hint

Common Mistakes

Dividing by max value only.

Dividing by min value.

Dividing by mean value.

Normalization requires dividing by the range: max - min.

4fill in blank

hard

Fill in the blank to create a dictionary of pixel depths for pixels with depth greater than 0.5.

Computer Vision

depth_pixels = {(x, y): depth_map[x, y] for x in range(depth_map.shape[0]) for y in range(depth_map.shape[1]) if depth_map[x, y] [1] 0.5}
print(len(depth_pixels))

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

A<=

B<

C>=

D>

Attempts:

3 left

💡 Hint

Common Mistakes

Using < instead of >.

Using <= instead of >=.

We want pixels with depth strictly greater than 0.5, so use >.

5fill in blank

hard

Fill all three blanks to compute mean squared error (MSE) between predicted and true depth maps.

Computer Vision

mse = sum((predicted_depth[1]true_depth)[2]2 for predicted_depth, true_depth in zip(predicted.flatten(), true.flatten())) / [3]
print(mse)

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

A-

B**

Clen(predicted.flatten())

D+

Attempts:

3 left

💡 Hint

Common Mistakes

Using addition instead of subtraction for difference.

Using single * instead of ** for power.

Dividing by wrong length.

MSE is calculated by squaring the difference (using - and **2) and dividing by the number of elements.