FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2994336323> ?p ?o ?g. }

• W2994336323 endingPage "8634" @default.
• W2994336323 startingPage "8615" @default.
• W2994336323 abstract "In recent years, long short-term memory neural networks (LSTMs) have been applied quite successfully to problems in handwritten text recognition. However, their strength is more located in handling sequences of variable length than in handling geometric variability of the image patterns. Furthermore, the best results for LSTMs are often based on large-scale training of an ensemble of network instances. In this paper, an end-to-end convolutional LSTM Neural Network is used to handle both geometric variation and sequence variability. We show that high performances can be reached on a common benchmark set by using proper data augmentation for just five such networks using a proper coding scheme and a proper voting scheme. The networks have similar architectures (Convolutional Neural Network (CNN): five layers, bidirectional LSTM (BiLSTM): three layers followed by a connectionist temporal classification (CTC) processing step). The approach assumes differently-scaled input images and different feature map sizes. Two datasets are used for evaluation of the performance of our algorithm: A standard benchmark RIMES dataset (French), and a historical handwritten dataset KdK (Dutch). Final performance obtained for the word-recognition test of RIMES was 96.6%, a clear improvement over other state-of-the-art approaches. On the KdK dataset, our approach also shows good results. The proposed approach is deployed in the Monk search engine for historical-handwriting collections." @default.
• W2994336323 created "2019-12-13" @default.
• W2994336323 creator A5028858025 @default.
• W2994336323 creator A5057913691 @default.
• W2994336323 date "2021-02-01" @default.
• W2994336323 modified "2023-09-26" @default.
• W2994336323 title "A limited-size ensemble of homogeneous CNN/LSTMs for high-performance word classification" @default.
• W2994336323 cites W146564573 @default.
• W2994336323 cites W1566256432 @default.
• W2994336323 cites W1585529631 @default.
• W2994336323 cites W1750919668 @default.
• W2994336323 cites W1884260092 @default.
• W2994336323 cites W1969698078 @default.
• W2994336323 cites W1970026646 @default.
• W2994336323 cites W1971938069 @default.
• W2994336323 cites W1973408529 @default.
• W2994336323 cites W1982849248 @default.
• W2994336323 cites W1993155451 @default.
• W2994336323 cites W1995574284 @default.
• W2994336323 cites W2005805506 @default.
• W2994336323 cites W2011468887 @default.
• W2994336323 cites W2018970719 @default.
• W2994336323 cites W2020103094 @default.
• W2994336323 cites W2025124369 @default.
• W2994336323 cites W2053317383 @default.
• W2994336323 cites W2057281839 @default.
• W2994336323 cites W2061994628 @default.
• W2994336323 cites W2064675550 @default.
• W2994336323 cites W2065546971 @default.
• W2994336323 cites W2066102695 @default.
• W2994336323 cites W2079145130 @default.
• W2994336323 cites W2096695793 @default.
• W2994336323 cites W2100235303 @default.
• W2994336323 cites W2102734279 @default.
• W2994336323 cites W2102834748 @default.
• W2994336323 cites W2106132157 @default.
• W2994336323 cites W2106843299 @default.
• W2994336323 cites W2110025586 @default.
• W2994336323 cites W2115094893 @default.
• W2994336323 cites W2117682730 @default.
• W2994336323 cites W2119615570 @default.
• W2994336323 cites W2123067645 @default.
• W2994336323 cites W2127141656 @default.
• W2994336323 cites W2130451799 @default.
• W2994336323 cites W2133463914 @default.
• W2994336323 cites W2142069714 @default.
• W2994336323 cites W2144499799 @default.
• W2994336323 cites W2149199613 @default.
• W2994336323 cites W2152928267 @default.
• W2994336323 cites W2152974942 @default.
• W2994336323 cites W2161969291 @default.
• W2994336323 cites W2181170718 @default.
• W2994336323 cites W2194187530 @default.
• W2994336323 cites W2263232528 @default.
• W2994336323 cites W2271075250 @default.
• W2994336323 cites W2328840452 @default.
• W2994336323 cites W2481930807 @default.
• W2994336323 cites W2499408409 @default.
• W2994336323 cites W2556301080 @default.
• W2994336323 cites W2573601984 @default.
• W2994336323 cites W2603013918 @default.
• W2994336323 cites W2725546079 @default.
• W2994336323 cites W2737677090 @default.
• W2994336323 cites W2787260200 @default.
• W2994336323 cites W2793404167 @default.
• W2994336323 cites W2796407073 @default.
• W2994336323 cites W2809598685 @default.
• W2994336323 cites W2904714269 @default.
• W2994336323 cites W2904970205 @default.
• W2994336323 cites W2906383034 @default.
• W2994336323 cites W2916222163 @default.
• W2994336323 cites W2963942586 @default.
• W2994336323 cites W2993383518 @default.
• W2994336323 cites W4312290170 @default.
• W2994336323 doi "https://doi.org/10.1007/s00521-020-05612-0" @default.
• W2994336323 hasPublicationYear "2021" @default.
• W2994336323 type Work @default.
• W2994336323 sameAs 2994336323 @default.
• W2994336323 citedByCount "6" @default.
• W2994336323 countsByYear W29943363232020 @default.
• W2994336323 countsByYear W29943363232022 @default.
• W2994336323 countsByYear W29943363232023 @default.
• W2994336323 crossrefType "journal-article" @default.
• W2994336323 hasAuthorship W2994336323A5028858025 @default.
• W2994336323 hasAuthorship W2994336323A5057913691 @default.
• W2994336323 hasBestOaLocation W29943363231 @default.
• W2994336323 hasConcept C13280743 @default.
• W2994336323 hasConcept C138885662 @default.
• W2994336323 hasConcept C153180895 @default.
• W2994336323 hasConcept C154945302 @default.
• W2994336323 hasConcept C185798385 @default.
• W2994336323 hasConcept C205649164 @default.
• W2994336323 hasConcept C2776401178 @default.
• W2994336323 hasConcept C2779386606 @default.
• W2994336323 hasConcept C41008148 @default.
• W2994336323 hasConcept C41895202 @default.
• W2994336323 hasConcept C50644808 @default.
• W2994336323 hasConcept C81363708 @default.

• next