Centre de documentation
scientifique

Label embedding : a frugal baseline for text recognition / Jose A. Rodriguez-Serrano in International journal of computer vision, vol 113 n° 3 (July 2015)  

Public [article]  inInternational journal of computer vision > vol 113 n° 3 (July 2015) . - pp 193 - 207 Titre : Label embedding : a frugal baseline for text recognition Type de document : Article/Communication Auteurs : Jose A. Rodriguez-Serrano, Auteur ; Albert Gordo, Auteur ; Florent Perronnin, Auteur Année de publication : 2015 Article en page(s) : pp 193 - 207 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Géomatique [Termes IGN] image [Termes IGN] reconnaissance de caractères [Termes IGN] segmentation sémantique [Termes IGN] test de performance Résumé : (Auteur) The standard approach to recognizing text in images consists in first classifying local image regions into candidate characters and then combining them with high-level word models such as conditional random fields. This paper explores a new paradigm that departs from this bottom-up view. We propose to embed word labels and word images into a common Euclidean space. Given a word image to be recognized, the text recognition problem is cast as one of retrieval: find the closest word label in this space. This common space is learned using the Structured SVM framework by enforcing matching label-image pairs to be closer than non-matching pairs. This method presents several advantages: it does not require ad-hoc or costly pre-/post-processing operations, it can build on top of any state-of-the-art image descriptor (Fisher vectors in our case), it allows for the recognition of never-seen-before words (zero-shot recognition) and the recognition process is simple and efficient, as it amounts to a nearest neighbor search. Experiments are performed on challenging datasets of license plates and scene text. The main conclusion of the paper is that with such a frugal approach it is possible to obtain results which are competitive with standard bottom-up approaches, thus establishing label embedding as an interesting and simple to compute baseline for text recognition. Numéro de notice : A2015--099 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Article DOI : 10.1007%2Fs11263-014-0793-6 En ligne : https://doi.org/10.1007/s11263-014-0793-6 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=85865 [article]