Matches in SemOpenAlex for { <https://semopenalex.org/work/W3197461909> ?p ?o ?g. }
- W3197461909 endingPage "316" @default.
- W3197461909 startingPage "290" @default.
- W3197461909 abstract "• We realize distorted target recognition by presenting a CapsNetSIFT architecture (cf. Fig. 1). It can simultaneously enable accurate positioning of interest regions and comprehensive learning of discriminative features whilst boasting invariance to visual distortions. • We propose a parallel MD-CapsNet in CapsNetSIFT for boosting representability. The customized structure, hyperparameter setting, and dynamic routing agreement have been demonstrated. • We propose a VM-SIFT in CapsNetSIFT to establish correspondence among capsule encoding vectors of standard images and distorted ones. This may represent the first attempt to realize feature matching of capsule vectors. • We conduct quantitative experiments of distorted target categorization on four benchmarks (CUB-200-2011, Stanford Dogs, Stanford Cars, and our hand-crafted rice growth dataset). Evaluation results reveal our advantage over state-of-the-arts. Due to overexposure, jitter, motion, and other spatiotemporal-varying perturbations, the collected images always undergo various visual distortions (e.g., deformation, partially occluded signs, fisheye respective, affine or 3D projections, in-plane and out-of-plane rotation) during acquisition or transmission procedure. Deep neural networks (DNNs) perform poorly on such pristine images in terms of high-level abstract operations, e.g., object categorization and semantic segmentation. To conquer this legacy, a distortion-tolerant model denoted as CapsNetSIFT is proposed to enhance representability and detectability of target in distorted imagery. We modify and integrate capsule network (CapsNet) with scale invariant feature transform (SIFT) together, both of which boast innate invariance to spacial-scale transformations. Two key insights, the customized multi-dimensional CapsNet (MD-CapsNet) and vector matching SIFT (VM-SIFT), can cooperate together and reinforce each other: the former encodes and provides representative feature vectors for the later, whilst the later localizes space-scale invariant interval dimensions (instead of pixels) and establish correspondence between source standard images (high-quality training images) and distorted ones (testing images). Thus, the category of one source standard image owning the most associations is the ground-truth category. Evaluation results reveal that employing CapsNetSIFT for distorted target recognition (CUB-200–2011, Stanford Dogs, Stanford Cars, and our hand-crafted dataset), significantly improves the resistance against various simulated distortions, and outperforms state-of-the-arts with relatively higher training and testing accuracy (93.97% and 91.03%)." @default.
- W3197461909 created "2021-09-13" @default.
- W3197461909 creator A5013019306 @default.
- W3197461909 creator A5038344148 @default.
- W3197461909 creator A5039297893 @default.
- W3197461909 creator A5078092833 @default.
- W3197461909 date "2021-11-01" @default.
- W3197461909 modified "2023-10-15" @default.
- W3197461909 title "CapsNet meets SIFT: A robust framework for distorted target categorization" @default.
- W3197461909 cites W1541110658 @default.
- W3197461909 cites W1565312575 @default.
- W3197461909 cites W1622676895 @default.
- W3197461909 cites W1677409904 @default.
- W3197461909 cites W1947161492 @default.
- W3197461909 cites W1965205435 @default.
- W3197461909 cites W1991367009 @default.
- W3197461909 cites W1995771589 @default.
- W3197461909 cites W2027324596 @default.
- W3197461909 cites W2034711024 @default.
- W3197461909 cites W2066624635 @default.
- W3197461909 cites W2074761716 @default.
- W3197461909 cites W2076890476 @default.
- W3197461909 cites W2088049833 @default.
- W3197461909 cites W2097117768 @default.
- W3197461909 cites W2118155754 @default.
- W3197461909 cites W2126291451 @default.
- W3197461909 cites W2130676164 @default.
- W3197461909 cites W2137405310 @default.
- W3197461909 cites W2138011018 @default.
- W3197461909 cites W2139337153 @default.
- W3197461909 cites W2151103935 @default.
- W3197461909 cites W2168588323 @default.
- W3197461909 cites W2194775991 @default.
- W3197461909 cites W2202499615 @default.
- W3197461909 cites W2220930568 @default.
- W3197461909 cites W2462457117 @default.
- W3197461909 cites W2466606639 @default.
- W3197461909 cites W2533598788 @default.
- W3197461909 cites W2578937925 @default.
- W3197461909 cites W2618530766 @default.
- W3197461909 cites W2778979395 @default.
- W3197461909 cites W2795330220 @default.
- W3197461909 cites W2798412221 @default.
- W3197461909 cites W2895526696 @default.
- W3197461909 cites W2910363199 @default.
- W3197461909 cites W2911968856 @default.
- W3197461909 cites W2919352650 @default.
- W3197461909 cites W2944401411 @default.
- W3197461909 cites W2944653015 @default.
- W3197461909 cites W2959131783 @default.
- W3197461909 cites W2962924597 @default.
- W3197461909 cites W2963314614 @default.
- W3197461909 cites W2963384288 @default.
- W3197461909 cites W2963583471 @default.
- W3197461909 cites W2963669006 @default.
- W3197461909 cites W2963967766 @default.
- W3197461909 cites W2963980515 @default.
- W3197461909 cites W2964237243 @default.
- W3197461909 cites W2979543786 @default.
- W3197461909 cites W2981510929 @default.
- W3197461909 cites W2991090397 @default.
- W3197461909 cites W2994144150 @default.
- W3197461909 cites W3033059362 @default.
- W3197461909 cites W3034302278 @default.
- W3197461909 cites W3035719652 @default.
- W3197461909 cites W3041089677 @default.
- W3197461909 doi "https://doi.org/10.1016/j.neucom.2021.08.087" @default.
- W3197461909 hasPublicationYear "2021" @default.
- W3197461909 type Work @default.
- W3197461909 sameAs 3197461909 @default.
- W3197461909 citedByCount "4" @default.
- W3197461909 countsByYear W31974619092021 @default.
- W3197461909 countsByYear W31974619092022 @default.
- W3197461909 countsByYear W31974619092023 @default.
- W3197461909 crossrefType "journal-article" @default.
- W3197461909 hasAuthorship W3197461909A5013019306 @default.
- W3197461909 hasAuthorship W3197461909A5038344148 @default.
- W3197461909 hasAuthorship W3197461909A5039297893 @default.
- W3197461909 hasAuthorship W3197461909A5078092833 @default.
- W3197461909 hasConcept C115961682 @default.
- W3197461909 hasConcept C153180895 @default.
- W3197461909 hasConcept C154945302 @default.
- W3197461909 hasConcept C31972630 @default.
- W3197461909 hasConcept C41008148 @default.
- W3197461909 hasConcept C61265191 @default.
- W3197461909 hasConcept C94124525 @default.
- W3197461909 hasConceptScore W3197461909C115961682 @default.
- W3197461909 hasConceptScore W3197461909C153180895 @default.
- W3197461909 hasConceptScore W3197461909C154945302 @default.
- W3197461909 hasConceptScore W3197461909C31972630 @default.
- W3197461909 hasConceptScore W3197461909C41008148 @default.
- W3197461909 hasConceptScore W3197461909C61265191 @default.
- W3197461909 hasConceptScore W3197461909C94124525 @default.
- W3197461909 hasFunder F4320313895 @default.
- W3197461909 hasFunder F4320321001 @default.
- W3197461909 hasFunder F4320321106 @default.
- W3197461909 hasFunder F4320324174 @default.
- W3197461909 hasFunder F4320335777 @default.