Object tracking using compressive local appearance model with ℓ1-regularisation

A novel compressive local appearance model-based object tracking algorithm is presented to address challenging issues in object tracking. To efficiently preserve image patches of an object and reduce the dimensionality, a random projection-based feature selection method is introduced. Modelling the...

Celý popis

Uložené v:

Podrobná bibliografia
Vydané v:	Electronics letters Ročník 50; číslo 6; s. 444 - 446
Hlavní autori:	Kim, Hyuncheol, Paik, Joonki
Médium:	Journal Article
Jazyk:	English
Vydavateľské údaje:	Stevenage The Institution of Engineering and Technology 13.03.2014 Institution of Engineering and Technology
Predmet:	Applied sciences Artificial intelligence Computer science; control theory; systems Data processing. List processing. Character string processing Detection, estimation, filtering, equalization, prediction dimensionality reduction Exact sciences and technology feature selection Image and vision processing and display technology image patches image representation Information, signal and communications theory matrix algebra Memory organisation. Data processing modified likelihood function novel compressive local appearance model‐based object tracking algorithm object tracking Pattern recognition. Digital image processing. Computational geometry random projection‐based feature selection method reconstruction error residual matrix Signal and communications theory Signal, noise Software sparse representation Telecommunications and information theory â„“1‐regularisation problem novel compressive local appearance model-based object tracking algorithm modified likelihood function residual matrix matrix algebra reconstruction error dimensionality reduction random projection-based feature selection method â„“1-regularisation problem image representation object tracking image patches feature selection sparse representation Occlusion Computer vision Data analysis Dimensionality Target tracking Tracking Image processing Local theory Long term Modeling Projection method Dimension reduction Experimental result Selection criterion Sparse representation Occultation Likelihood function
ISSN:	0013-5194, 1350-911X, 1350-911X
On-line prístup:	Získať plný text
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!

Popis
Shrnutí:	A novel compressive local appearance model-based object tracking algorithm is presented to address challenging issues in object tracking. To efficiently preserve image patches of an object and reduce the dimensionality, a random projection-based feature selection method is introduced. Modelling the object's appearance using a sparse representation over a set of templates leads to an ℓ1-regularisation problem. To solve this problem, both the reconstruction error and the residual matrix are considered which play a key role in tracking an object with severe appearance variations using the modified likelihood function. Experimental results demonstrate that the proposed method outperforms existing state-of-the-art tracking methods in terms of dealing with long-term partial occlusion, deformation and rotation.
ISSN:	0013-5194 1350-911X 1350-911X
DOI:	10.1049/el.2013.2763