Mostrando 5 resultados de: 5
Filtros aplicados
Publisher
Pattern Recognition(2)
Proceedings of the International Conference on Document Analysis and Recognition, ICDAR(2)
Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017(1)
Área temáticas
Funcionamiento de bibliotecas y archivos(4)
Métodos informáticos especiales(4)
Ciencias de la computación(1)
Doctrinas(1)
Educación de adultos(1)
Área de conocimiento
Ciencias de la computación(4)
Aprendizaje automático(2)
Visión por computadora(2)
Algoritmo(1)
Objetivos de Desarrollo Sostenible
ODS 4: Educación de calidad(4)
ODS 9: Industria, innovación e infraestructura(4)
ODS 17: Alianzas para lograr los objetivos(3)
ODS 11: Ciudades y comunidades sostenibles(1)
Origen
scopus(5)
ICDAR2017 Robust Reading Challenge on Omnidirectional Video
Conference ObjectAbstract: Results of ICDAR 2017 Robust Reading Challenge on Omnidirectional Video are presented. This competitPalabras claves:DOST, Downtown Osaka Scene Text Dataset, Japanese, Multi-scipt text, Omnidirectional video, Robust readingAutores:Bazazian D., Iwamura M., Karatzas D., Lluís Álvarez Gómez, Morimoto N., Tainaka K.Fuentes:scopusImproving patch-based scene text script identification with ensembles of conjoined networks
ArticleAbstract: This paper focuses on the problem of script identification in scene text images. Facing this problemPalabras claves:convolutional neural networks, Ensemble of conjoined networks, Multi-language OCR, Scene text understanding, script identificationAutores:Karatzas D., Lluís Álvarez Gómez, Nicolaou A.Fuentes:scopusLSDE: Levenshtein Space Deep Embedding for Query-by-String Word Spotting
Conference ObjectAbstract: In this paper we present the LSDE string representation and its application to handwritten word spotPalabras claves:CNNs, Deep embeddings, Handwritten Keyword Spotting, Query by stringAutores:Karatzas D., Lluís Álvarez Gómez, Rusiñol M.Fuentes:scopusSelf-supervised learning of visual features through embedding images into text topic spaces
Conference ObjectAbstract: End-to-end training from scratch of current deep architectures for new computer vision problems woulPalabras claves:Autores:Jawahar C.V., Karatzas D., Lluís Álvarez Gómez, Patel Y., Rusiñol M.Fuentes:scopusTextProposals: A text-specific selective search algorithm for word spotting in the wild
ArticleAbstract: Motivated by the success of powerful while expensive techniques to recognize words in a holistic wayPalabras claves:Grouping, Object proposals, Perceptual organization, Scene textAutores:Karatzas D., Lluís Álvarez GómezFuentes:scopus