Feature matching between images in Computer Vision - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is feature matching between images?

Feature matching is the process of finding corresponding points or patterns between two or more images. It helps computers understand how images relate to each other, like matching puzzle pieces.

beginner

Name two common feature detectors used in image matching.

Two common feature detectors are SIFT (Scale-Invariant Feature Transform) and ORB (Oriented FAST and Rotated BRIEF). They find key points in images that are easy to match.

beginner

Why do we use descriptors in feature matching?

Descriptors describe the area around a key point with numbers. This helps compare features between images by turning visual patterns into data that computers can match.

intermediate

What is the role of the RANSAC algorithm in feature matching?

RANSAC helps find the best matches by ignoring wrong matches (outliers). It fits a model that explains most matches, making the matching more accurate.

intermediate

Explain the difference between brute-force matching and FLANN matching.

Brute-force matching compares every feature in one image to every feature in another, which is simple but slow. FLANN (Fast Library for Approximate Nearest Neighbors) uses smart searching to find matches faster.

Which of these is a feature detector used in image matching?

A ReLU

B SIFT

C Dropout

D BatchNorm

SIFT is a feature detector that finds key points in images. ReLU, Dropout, and BatchNorm are used in neural networks.

What does a descriptor do in feature matching?

A Describes the area around a key point numerically

B Converts images to grayscale

C Removes noise from images

D Detects edges in images

Descriptors turn the area around key points into numbers to help compare features between images.

Why is RANSAC used in feature matching?

A To increase image resolution

B To speed up feature detection

C To convert images to binary

D To remove incorrect matches

RANSAC removes wrong matches (outliers) to improve the accuracy of matching.

Which matching method compares every feature with every other feature?

A Brute-force

B RANSAC

C FLANN

D SIFT

Brute-force matching compares all features directly, which is simple but can be slow.

What is the main advantage of FLANN over brute-force matching?

A Detects more features

B More accurate matches

C Faster search for matches

D Removes noise automatically

FLANN uses approximate nearest neighbor search to find matches faster than brute-force.

Describe the steps involved in feature matching between two images.

Think about how you find and compare puzzle pieces.

Explain why feature matching is important in computer vision applications.

Consider how matching helps connect different views of the same scene.