ORB features in Computer Vision - Interactive Code Practice

Practice - 5 Tasks

1fill in blank

easy

Complete the code to create an ORB detector using OpenCV.

Computer Vision

import cv2
orb = cv2.[1]()

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

AFAST_create

BSIFT_create

CBRISK_create

DORB_create

Attempts:

3 left

💡 Hint

Common Mistakes

Using SIFT_create instead of ORB_create

Using FAST_create which is a different detector

Trying to instantiate ORB directly without the create method

The correct function to create an ORB detector in OpenCV is ORB_create().

2fill in blank

medium

Complete the code to detect keypoints and compute descriptors from an image using ORB.

Computer Vision

keypoints, descriptors = orb.[1](image)

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

Adetect

Bcompute

CdetectAndCompute

DfindKeypoints

Attempts:

3 left

💡 Hint

Common Mistakes

Using detect only, which returns keypoints but no descriptors

Using compute only, which requires keypoints as input

Using a non-existent method like findKeypoints

The detectAndCompute method detects keypoints and computes descriptors in one step.

3fill in blank

hard

Fix the error in the code to correctly draw keypoints on the image.

Computer Vision

output_image = cv2.drawKeypoints(image, keypoints, [1], flags=cv2.DRAW_MATCHES_FLAGS_DRAW_RICH_KEYPOINTS)

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

Akeypoints

BNone

Coutput_image

Dimage

Attempts:

3 left

💡 Hint

Common Mistakes

Passing the input image as the output image argument

Passing keypoints instead of an image

Passing an uninitialized variable

The third argument to drawKeypoints should be None to create a new output image with keypoints drawn.

4fill in blank

hard

Fill both blanks to create a dictionary of ORB keypoints and their sizes for keypoints larger than 20.

Computer Vision

keypoint_sizes = {kp.pt: kp.[1] for kp in keypoints if kp.[1] [2] 20}

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

Asize

Bangle

C>

D<

Attempts:

3 left

💡 Hint

Common Mistakes

Using 'angle' instead of 'size' for the attribute

Using '<' instead of '>' for filtering

Trying to access non-existent attributes

We access the size attribute of each keypoint and filter those with size greater than 20.

5fill in blank

hard

Fill all three blanks to create a dictionary of keypoint coordinates and their response values for keypoints with response above 0.01.

Computer Vision

keypoint_responses = {(int(kp.pt[0]), int(kp.pt[1])): kp.[1] for kp in keypoints if kp.[2] [3] 0.01}

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

Aresponse

C>

D<

Attempts:

3 left

💡 Hint

Common Mistakes

Using '<' instead of '>' for filtering

Using wrong attribute names

Not converting coordinates to integers

The response attribute measures the strength of a keypoint. We filter keypoints with response greater than 0.01.