Kanade–Lucas–Tomasi feature tracker

E295650

computer vision algorithm feature tracking algorithm optical flow-based method

The Kanade–Lucas–Tomasi feature tracker is a widely used computer vision algorithm for robustly tracking distinctive image features across video frames, building on the Lucas–Kanade optical flow method with Tomasi’s feature selection criteria.

Try in SPARQL Jump to: Surface forms Disambiguation Statements Elicitation Referenced by

All labels observed (3)

Label	Occurrences
KLT feature tracker	1
Kanade–Lucas–Tomasi feature tracker canonical	1
Tomasi–Kanade feature selection criterion	1

How this entity was disambiguated

This entity first appeared as the object of triple T2739075 — resolving that mention is where its identity was fixed. The disambiguator weighed these candidate entities and picked the highlighted one (or “None”, minting a new entity). This is how homonymy is resolved: the same surface form can point to different entities.

NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07

Target entity: Kanade–Lucas–Tomasi feature tracker
Context triple: [Takeo Kanade, knownFor, Kanade–Lucas–Tomasi feature tracker]

A. SLAM
SLAM is a major art museum in St. Louis, Missouri, renowned for its extensive collection spanning thousands of years and diverse cultures.
B. ANSAC
ANSAC is the abbreviation for the Applied and Natural Science Accreditation Commission, a body that accredits applied and natural science degree programs.
C. TCSVT
TCSVT is a leading IEEE journal that publishes research on circuits, systems, and technologies for video and multimedia processing.
D. IEEE Computer Society Conference on Computer Vision and Pattern Recognition
The IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR) is a premier annual international research conference showcasing cutting-edge advances in computer vision, machine learning, and pattern recognition.
E. Kalman filter
The Kalman filter is a mathematical algorithm used to estimate the changing state of a system from noisy measurements, widely applied in control systems, navigation, and signal processing.
F. None of above. chosen
G. Unsure - the case is ambiguous/there is not enough information to decide.

NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07

Target entity: Kanade–Lucas–Tomasi feature tracker
Target entity description: The Kanade–Lucas–Tomasi feature tracker is a widely used computer vision algorithm for robustly tracking distinctive image features across video frames, building on the Lucas–Kanade optical flow method with Tomasi’s feature selection criteria.

A. SLAM
SLAM is a major art museum in St. Louis, Missouri, renowned for its extensive collection spanning thousands of years and diverse cultures.
B. ANSAC
ANSAC is the abbreviation for the Applied and Natural Science Accreditation Commission, a body that accredits applied and natural science degree programs.
C. TCSVT
TCSVT is a leading IEEE journal that publishes research on circuits, systems, and technologies for video and multimedia processing.
D. IEEE Computer Society Conference on Computer Vision and Pattern Recognition
The IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR) is a premier annual international research conference showcasing cutting-edge advances in computer vision, machine learning, and pattern recognition.
E. Kalman filter
The Kalman filter is a mathematical algorithm used to estimate the changing state of a system from noisy measurements, widely applied in control systems, navigation, and signal processing.
F. None of above. chosen

Statements (46)

Predicate	Object
instanceOf	computer vision algorithm ⓘ feature tracking algorithm ⓘ optical flow-based method ⓘ
assumes	brightness constancy ⓘ small motion between frames ⓘ spatial coherence of motion ⓘ
basedOn	Lucas–Kanade optical flow algorithm ⓘ surface form: Lucas–Kanade optical flow method
category	feature detection and tracking ⓘ optical flow ⓘ
featureSelectionCriterion	eigenvalue-based corner strength ⓘ good features to track ⓘ
field	computer vision ⓘ image processing ⓘ
input	image sequences ⓘ video frames ⓘ
limitation	difficulty with large inter-frame motion without pyramids ⓘ sensitive to strong appearance changes ⓘ
oftenImplementedWith	image pyramids ⓘ multi-scale tracking ⓘ
output	trajectories of tracked features ⓘ
property	computationally efficient ⓘ robust to moderate noise ⓘ suitable for real-time applications ⓘ
relatedTo	Harris corner detector ⓘ Shi–Tomasi corner detector ⓘ Lucas–Kanade optical flow algorithm ⓘ surface form: pyramidal Lucas–Kanade method
requires	good spatial texture around feature points ⓘ
robustness	handles moderate illumination changes ⓘ handles partial occlusions of features ⓘ
selectsFeaturesWith	large minimum eigenvalue of gradient covariance matrix ⓘ
task	feature tracking ⓘ motion tracking in image sequences ⓘ
tracks	distinctive image features ⓘ interest points ⓘ
usedBy	OpenCV ⓘ surface form: OpenCV library many real-time vision systems ⓘ
usedIn	SLAM ⓘ augmented reality ⓘ object tracking ⓘ structure from motion ⓘ video stabilization ⓘ visual odometry ⓘ
uses	Kanade–Lucas–Tomasi feature tracker self-linksurface differs ⓘ surface form: Tomasi–Kanade feature selection criterion gradient-based optimization ⓘ local image patch alignment ⓘ sparse optical flow ⓘ

How these facts were elicited

Referenced by (3)

Full triples — surface form annotated when it differs from this entity's canonical label.

Takeo Kanade → knownFor → Kanade–Lucas–Tomasi feature tracker ⓘ

Lucas–Kanade optical flow algorithm → relatedTo → Kanade–Lucas–Tomasi feature tracker ⓘ

this entity surface form: KLT feature tracker

Kanade–Lucas–Tomasi feature tracker → uses → Kanade–Lucas–Tomasi feature tracker self-linksurface differs ⓘ

this entity surface form: Tomasi–Kanade feature selection criterion

All labels observed (3)

How this entity was disambiguated Show

Statements (46)

How these facts were elicited Show

Referenced by (3)

How this entity was disambiguated

How these facts were elicited