Why features identify distinctive points in Computer Vision - Test Your Understanding

Practice - 5 Tasks

1fill in blank

easy

Complete the code to extract keypoints using ORB detector.

Computer Vision

import cv2
img = cv2.imread('image.jpg', 0)
orb = cv2.ORB_create()
keypoints = orb.[1](img)

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

Aextract

BfindFeatures

Cdetect

DdetectAndCompute

Attempts:

3 left

💡 Hint

Common Mistakes

Using detectAndCompute instead of detect

Trying to use non-existent methods like findFeatures

The detect method finds keypoints in the image, which are distinctive points.

2fill in blank

medium

Complete the code to compute descriptors for the detected keypoints.

Computer Vision

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

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

Aextract

BdetectAndCompute

Ccompute

Ddetect

Attempts:

3 left

💡 Hint

Common Mistakes

Using compute which requires keypoints as input and returns descriptors only

Using detect which returns only keypoints

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

3fill in blank

hard

Fix the error in the code to correctly match descriptors using BFMatcher.

Computer Vision

bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)
matches = bf.[1](des1, des2)

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

AfindMatches

Bmatches

CknnMatch

Dmatch

Attempts:

3 left

💡 Hint

Common Mistakes

Using matches which is not a method

Using knnMatch when only single matches are needed

The match method finds the best matches between two sets of descriptors.

4fill in blank

hard

Fill both blanks to create a dictionary of keypoints and their coordinates.

Computer Vision

kp_dict = {kp.[1]: (kp.[2][0], kp.[2][1]) for kp in keypoints}

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

Aangle

Bpt

Csize

Dresponse

Attempts:

3 left

💡 Hint

Common Mistakes

Using angle or size which are not coordinates

Trying to use response which is a strength measure

The pt attribute of a keypoint gives its (x, y) coordinates.

5fill in blank

hard

Fill all three blanks to filter good matches based on distance.

Computer Vision

good_matches = [m for m in matches if m.[1] < [2] * min_dist and m.[3] != 0]

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

Adistance

B0.7

CqueryIdx

DtrainIdx

Attempts:

3 left

💡 Hint

Common Mistakes

Using trainIdx instead of queryIdx

Using wrong threshold values

Checking wrong attributes

We filter matches where the distance is less than 0.7 times the minimum distance, and the queryIdx is not zero to avoid invalid matches.