FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2887556118> ?p ?o ?g. }

• W2887556118 endingPage "253" @default.
• W2887556118 startingPage "240" @default.
• W2887556118 abstract "Visual tracking algorithms based on structured output support vector machine (SOSVM) have demonstrated excellent performance. However, sampling methods and optimization strategies of SOSVM undesirably increase the computational overloads, which hinder real-time application of these algorithms. Moreover, due to the lack of high-dimensional features and dense training samples, SOSVM-based algorithms are unstable to deal with various challenging scenarios, such as occlusions and scale variations. Recently, visual tracking algorithms based on discriminative correlation filters (DCF), especially the combination of DCF and features from deep convolutional neural networks (CNN), have been successfully applied to visual tracking, and attains surprisingly good performance on recent benchmarks. The success is mainly attributed to two aspects: the circular correlation properties of DCF and the powerful representation capabilities of CNN features. Nevertheless, compared with SOSVM, DCF-based algorithms are restricted to simple ridge regression which has a weaker discriminative ability. In this paper, a novel circular and structural operator tracker (CSOT) is proposed for high performance visual tracking, it not only possesses the powerful discriminative capability of SOSVM but also efficiently inherits the superior computational efficiency of DCF. Based on the proposed circular and structural operators, a set of primal confidence score maps can be obtained by circular correlating feature maps with their corresponding structural correlation filters. Furthermore, an implicit interpolation is applied to convert the multi-resolution feature maps to the continuous domain and make all primal confidence score maps have the same spatial resolution. Then, we exploit an efficient ensemble post-processor based on relative entropy, which can coalesce primal confidence score maps and create an optimal confidence score map for more accurate localization. The target is localized on the peak of the optimal confidence score map. Besides, we introduce a collaborative optimization strategy to update circular and structural operators by iteratively training structural correlation filters, which significantly reduces computational complexity and improves robustness. Experimental results demonstrate that our approach achieves state-of-the-art performance in mean AUC scores of 71.5% and 69.4% on the OTB2013 and OTB2015 benchmarks respectively, and obtains a third-best expected average overlap (EAO) score of 29.8% on the VOT2017 benchmark." @default.
• W2887556118 created "2018-08-22" @default.
• W2887556118 creator A5025527694 @default.
• W2887556118 creator A5028272399 @default.
• W2887556118 creator A5037225064 @default.
• W2887556118 creator A5051969162 @default.
• W2887556118 creator A5072675100 @default.
• W2887556118 creator A5086529957 @default.
• W2887556118 creator A5086664284 @default.
• W2887556118 date "2018-12-01" @default.
• W2887556118 modified "2023-10-17" @default.
• W2887556118 title "High performance visual tracking with circular and structural operators" @default.
• W2887556118 cites W161114242 @default.
• W2887556118 cites W182940129 @default.
• W2887556118 cites W1857884451 @default.
• W2887556118 cites W1904671147 @default.
• W2887556118 cites W1915599933 @default.
• W2887556118 cites W1923332106 @default.
• W2887556118 cites W1955514522 @default.
• W2887556118 cites W1963882359 @default.
• W2887556118 cites W1964846093 @default.
• W2887556118 cites W2027714080 @default.
• W2887556118 cites W2044986361 @default.
• W2887556118 cites W2072348815 @default.
• W2887556118 cites W2089961441 @default.
• W2887556118 cites W2098711149 @default.
• W2887556118 cites W2102605133 @default.
• W2887556118 cites W2118345389 @default.
• W2887556118 cites W2124211486 @default.
• W2887556118 cites W2126302311 @default.
• W2887556118 cites W2154889144 @default.
• W2887556118 cites W2157700924 @default.
• W2887556118 cites W2158592639 @default.
• W2887556118 cites W2158827467 @default.
• W2887556118 cites W2162349892 @default.
• W2887556118 cites W2162919312 @default.
• W2887556118 cites W2168356304 @default.
• W2887556118 cites W2244956674 @default.
• W2887556118 cites W2335586288 @default.
• W2887556118 cites W2343187456 @default.
• W2887556118 cites W2470394683 @default.
• W2887556118 cites W2474599091 @default.
• W2887556118 cites W2494699027 @default.
• W2887556118 cites W2518013266 @default.
• W2887556118 cites W2520477759 @default.
• W2887556118 cites W2557641257 @default.
• W2887556118 cites W2558899534 @default.
• W2887556118 cites W2576085163 @default.
• W2887556118 cites W2599547527 @default.
• W2887556118 cites W2609910506 @default.
• W2887556118 cites W2673818281 @default.
• W2887556118 cites W2681067697 @default.
• W2887556118 cites W2735270199 @default.
• W2887556118 cites W2737362155 @default.
• W2887556118 cites W2737572441 @default.
• W2887556118 cites W2768634781 @default.
• W2887556118 cites W2899321954 @default.
• W2887556118 cites W2916780012 @default.
• W2887556118 cites W2917435394 @default.
• W2887556118 cites W29474918 @default.
• W2887556118 cites W2962799386 @default.
• W2887556118 cites W2962864296 @default.
• W2887556118 cites W2962968160 @default.
• W2887556118 cites W2964111344 @default.
• W2887556118 cites W3102624093 @default.
• W2887556118 cites W3106250896 @default.
• W2887556118 cites W818325216 @default.
• W2887556118 doi "https://doi.org/10.1016/j.knosys.2018.08.008" @default.
• W2887556118 hasPublicationYear "2018" @default.
• W2887556118 type Work @default.
• W2887556118 sameAs 2887556118 @default.
• W2887556118 citedByCount "12" @default.
• W2887556118 countsByYear W28875561182018 @default.
• W2887556118 countsByYear W28875561182019 @default.
• W2887556118 countsByYear W28875561182020 @default.
• W2887556118 countsByYear W28875561182021 @default.
• W2887556118 countsByYear W28875561182023 @default.
• W2887556118 crossrefType "journal-article" @default.
• W2887556118 hasAuthorship W2887556118A5025527694 @default.
• W2887556118 hasAuthorship W2887556118A5028272399 @default.
• W2887556118 hasAuthorship W2887556118A5037225064 @default.
• W2887556118 hasAuthorship W2887556118A5051969162 @default.
• W2887556118 hasAuthorship W2887556118A5072675100 @default.
• W2887556118 hasAuthorship W2887556118A5086529957 @default.
• W2887556118 hasAuthorship W2887556118A5086664284 @default.
• W2887556118 hasBestOaLocation W28875561182 @default.
• W2887556118 hasConcept C11413529 @default.
• W2887556118 hasConcept C138885662 @default.
• W2887556118 hasConcept C153180895 @default.
• W2887556118 hasConcept C154945302 @default.
• W2887556118 hasConcept C2776401178 @default.
• W2887556118 hasConcept C41008148 @default.
• W2887556118 hasConcept C41895202 @default.
• W2887556118 hasConcept C56461940 @default.
• W2887556118 hasConcept C81363708 @default.
• W2887556118 hasConcept C97931131 @default.
• W2887556118 hasConceptScore W2887556118C11413529 @default.
• W2887556118 hasConceptScore W2887556118C138885662 @default.

• next