FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4311773086> ?p ?o ?g. }

• W4311773086 endingPage "109912" @default.
• W4311773086 startingPage "109912" @default.
• W4311773086 abstract "In recent years, Siamese-based trackers have shown remarkable improvement in visual tracking. The general trends are interested in making deeper and more complicated networks to pursue higher accuracy. However, these advances result in cumbersome trackers with respect to size and speed, which hinders the deployment of deep trackers on edge devices. Due to the tracking scenario complexity and temporal coherence, the backbone network of deep trackers emphasizes the target appearance information more. However, the appearance feature is sensitive to parameter variation, making the traditional deep model compression methods hard to compress a deep tracker. To bridge the gap between deep Siamese trackers and practical use, we propose a new feature distillation algorithm suitable for deep trackers in this paper. Firstly, motivated by the concept of divide-and-conquer, we formulate the feature distillation into a stepwise distillation problem and perform distillation on each minimum unit to relieve the hard-to-distill problem of appearance feature. Secondly, we reconstruct the student model into a combination of convolution kernels and a point-wise convolutional layer, which enables the student model to inherit all the parameters of the teacher model during initialization. Finally, we propose a 3-step warm-up training strategy to address the student model’s degradation and structural adaptation problems during training. Extensive experiments on eight benchmarks demonstrate that our proposed method compresses the fully convolutional Siamese Networks (SiamFC) and its variant Siamd and achieves leading tracking performance with only 2.1 MB model size while running at 225 fps." @default.
• W4311773086 created "2022-12-28" @default.
• W4311773086 creator A5025929169 @default.
• W4311773086 creator A5044544424 @default.
• W4311773086 creator A5052943292 @default.
• W4311773086 creator A5064143081 @default.
• W4311773086 date "2023-01-01" @default.
• W4311773086 modified "2023-10-11" @default.
• W4311773086 title "Feature distillation Siamese networks for object tracking" @default.
• W4311773086 cites W1677182931 @default.
• W4311773086 cites W1857884451 @default.
• W4311773086 cites W1915785815 @default.
• W4311773086 cites W1932847118 @default.
• W4311773086 cites W2062118960 @default.
• W4311773086 cites W2117539524 @default.
• W4311773086 cites W2154889144 @default.
• W4311773086 cites W2158592639 @default.
• W4311773086 cites W2194775991 @default.
• W4311773086 cites W2214352687 @default.
• W4311773086 cites W2408241409 @default.
• W4311773086 cites W2469582947 @default.
• W4311773086 cites W2470394683 @default.
• W4311773086 cites W2518013266 @default.
• W4311773086 cites W2518876086 @default.
• W4311773086 cites W2520477759 @default.
• W4311773086 cites W2554302513 @default.
• W4311773086 cites W2557641257 @default.
• W4311773086 cites W2558899534 @default.
• W4311773086 cites W2592463526 @default.
• W4311773086 cites W2599547527 @default.
• W4311773086 cites W2604701225 @default.
• W4311773086 cites W2618530766 @default.
• W4311773086 cites W2740685955 @default.
• W4311773086 cites W2772757469 @default.
• W4311773086 cites W2776035257 @default.
• W4311773086 cites W2794744029 @default.
• W4311773086 cites W2799058067 @default.
• W4311773086 cites W2884367402 @default.
• W4311773086 cites W2884561390 @default.
• W4311773086 cites W2886910176 @default.
• W4311773086 cites W2887240040 @default.
• W4311773086 cites W2891033863 @default.
• W4311773086 cites W2892241171 @default.
• W4311773086 cites W2898200825 @default.
• W4311773086 cites W2913466142 @default.
• W4311773086 cites W2955983623 @default.
• W4311773086 cites W2963227409 @default.
• W4311773086 cites W2963450965 @default.
• W4311773086 cites W2963534981 @default.
• W4311773086 cites W2963854930 @default.
• W4311773086 cites W2964069521 @default.
• W4311773086 cites W2964099559 @default.
• W4311773086 cites W2964111344 @default.
• W4311773086 cites W2964217848 @default.
• W4311773086 cites W2964242925 @default.
• W4311773086 cites W2964423614 @default.
• W4311773086 cites W2982157312 @default.
• W4311773086 cites W2987460522 @default.
• W4311773086 cites W2987906097 @default.
• W4311773086 cites W2996167212 @default.
• W4311773086 cites W3004543888 @default.
• W4311773086 cites W3012561096 @default.
• W4311773086 cites W3044714827 @default.
• W4311773086 cites W3092074970 @default.
• W4311773086 cites W3102624093 @default.
• W4311773086 cites W3180327660 @default.
• W4311773086 cites W4207024994 @default.
• W4311773086 cites W4312821902 @default.
• W4311773086 cites W823218635 @default.
• W4311773086 doi "https://doi.org/10.1016/j.asoc.2022.109912" @default.
• W4311773086 hasPublicationYear "2023" @default.
• W4311773086 type Work @default.
• W4311773086 citedByCount "3" @default.
• W4311773086 countsByYear W43117730862023 @default.
• W4311773086 crossrefType "journal-article" @default.
• W4311773086 hasAuthorship W4311773086A5025929169 @default.
• W4311773086 hasAuthorship W4311773086A5044544424 @default.
• W4311773086 hasAuthorship W4311773086A5052943292 @default.
• W4311773086 hasAuthorship W4311773086A5064143081 @default.
• W4311773086 hasConcept C108583219 @default.
• W4311773086 hasConcept C114466953 @default.
• W4311773086 hasConcept C138885662 @default.
• W4311773086 hasConcept C153180895 @default.
• W4311773086 hasConcept C154945302 @default.
• W4311773086 hasConcept C199360897 @default.
• W4311773086 hasConcept C2776401178 @default.
• W4311773086 hasConcept C31972630 @default.
• W4311773086 hasConcept C41008148 @default.
• W4311773086 hasConcept C41895202 @default.
• W4311773086 hasConcept C56461940 @default.
• W4311773086 hasConcept C57501372 @default.
• W4311773086 hasConceptScore W4311773086C108583219 @default.
• W4311773086 hasConceptScore W4311773086C114466953 @default.
• W4311773086 hasConceptScore W4311773086C138885662 @default.
• W4311773086 hasConceptScore W4311773086C153180895 @default.
• W4311773086 hasConceptScore W4311773086C154945302 @default.
• W4311773086 hasConceptScore W4311773086C199360897 @default.
• W4311773086 hasConceptScore W4311773086C2776401178 @default.

• next