Mostrando 8 resultados de: 8
Filtros aplicados
Subtipo de publicación
Conference Object(8)
Área temáticas
Métodos informáticos especiales(5)
Comunicaciones(1)
Funcionamiento de bibliotecas y archivos(1)
Instrumentos de precisión y otros dispositivos(1)
Interacción social(1)
Área de conocimiento
Ciencias de la computación(6)
Visión por computadora(4)
Análisis de datos(2)
Aprendizaje automático(1)
Idioma(1)
Origen
scopus(8)
Dynamic lexicon generation for natural scene images
Conference ObjectAbstract: Many scene text understanding methods approach the endto-end recognition problem from a word-spottinPalabras claves:cnn, Lexicon generation, Photo OCR, Scene text, Scene understanding, Topic modelingAutores:Karatzas D., Lluís Álvarez Gómez, Patel Y., Rusiñol M.Fuentes:scopusOCR-IDL: OCR Annotations for Industry Document Library Dataset
Conference ObjectAbstract: Pretraining has proven successful in Document Intelligence tasks where deluge of documents are usedPalabras claves:Autores:Furkan Biten A., Karatzas D., Lluís Álvarez Gómez, Tito R., Valveny E.Fuentes:scopusMUST-VQA: MUltilingual Scene-Text VQA
Conference ObjectAbstract: In this paper, we present a framework for Multilingual Scene Text Visual Question Answering that deaPalabras claves:Multilingual models, Power of language models, Scene text, Translation robustness, Visual question answering, Zero-shot transferAutores:Furkan Biten A., Karatzas D., Lluís Álvarez Gómez, Mafla A., Vivoli E.Fuentes:scopusLearning from #barcelona instagram data what locals and tourists post about its neighbourhoods
Conference ObjectAbstract: Massive tourism is becoming a big problem for some cities, such as Barcelona, due to its concentratiPalabras claves:City tourism analysis, Self-supervised learning, Social media analysis, Webly supervised learningAutores:Gibert J., Gómez R., Karatzas D., Lluís Álvarez GómezFuentes:scopusLearning to learn from web data through deep semantic embeddings
Conference ObjectAbstract: In this paper we propose to learn a multimodal image and text embedding from Web and Social Media daPalabras claves:Multimodal embeddings, Multimodal retrieval, Self-supervised learning, Text embeddings, Webly supervised learningAutores:Gibert J., Gómez R., Karatzas D., Lluís Álvarez GómezFuentes:scopusLocation Sensitive Image Retrieval and Tagging
Conference ObjectAbstract: People from different parts of the globe describe objects and concepts in distinct manners. Visual aPalabras claves:Autores:Gibert J., Gómez R., Karatzas D., Lluís Álvarez GómezFuentes:scopusScene text recognition: No country for old men?
Conference ObjectAbstract: It is a generally accepted fact that Off-the-shelf OCR engines do not perform well in unconstrainedPalabras claves:Autores:Karatzas D., Lluís Álvarez GómezFuentes:scopusSingle shot scene text retrieval
Conference ObjectAbstract: Textual information found in scene images provides high level semantic information about the image aPalabras claves:convolutional neural networks, Image retrieval, PHOC, Region proposals networks, Scene text, Word spottingAutores:Karatzas D., Lluís Álvarez Gómez, Mafla A., Rusiñol M.Fuentes:scopus