|
|
|||||||||||||||
|
Cerca | Lliura | Ajuda | Servei de Biblioteques | Sobre el DDD | Català English Español | |||||||||
| Pàgina inicial > Articles > Articles publicats > Real-time Lexicon-free Scene Text Retrieval |
| Data: | 2021 |
| Resum: | In this work, we address the task of scene text retrieval: given a text query, the system returns all im- ages containing the queried text. The proposed model uses a single shot CNN architecture that predicts bounding boxes and builds a compact representation of spotted words. In this way, this problem can be modeled as a nearest neighbor search of the textual representation of a query over the outputs of the CNN collected from the totality of an image database. Our experiments demonstrate that the proposed model outperforms previous state-of-the-art, while offering a significant increase in processing speed and unmatched expressiveness with samples never seen at training time. Several experiments to assess the generalization capability of the model are conducted in a multilingual dataset, as well as an application of real-time text spotting in videos. |
| Ajuts: | Agencia Estatal de Investigación TIN2017-89779-P Generalitat de Catalunya 2019/FI_B01233 European Commission 712949 |
| Nota: | Altres ajuts: European Social Fund 2014-2020 (CCI: 2014ES05SFOP007); CERCA Programme / Generalitat de Catalunya; UAB Ph.D. scholarship No B18P0070 |
| Drets: | Aquest document està subjecte a una llicència d'ús Creative Commons. Es permet la reproducció total o parcial, la distribució, i la comunicació pública de l'obra, sempre que no sigui amb finalitats comercials, i sempre que es reconegui l'autoria de l'obra original. No es permet la creació d'obres derivades. |
| Llengua: | Anglès |
| Document: | Article ; recerca ; Versió acceptada per publicar |
| Matèria: | Convolutional neural networks ; Image retrieval ; PHOC ; Region proposal networks ; Scene text detection ; Scene text recognition ; Word spotting |
| Publicat a: | Pattern Recognition, Vol. 110 (February 2021) , art. 107656, ISSN 0031-3203 |
Postprint 31 p, 5.0 MB |