1,721,013 research outputs found
Les droits linguistiques en Europe et ailleurs - Linguistic Rights: Europe and Beyond, Atti delle Prime Giornate dei Diritti Linguistici (Università di Teramo, 11-12 giugno 2007)
No full textIl motivo del manoscritto genetico nei romanzi arturiani francesi in versi (1150-1250) / 2-3
No full text
Produzione culturale ed emancipazione delle minoranze linguistiche. L’esempio delle isole alloglotte francoprovenzali di Faeto e Celle di San Vito (FG)
No full text
L'enseignement des langues locales : institutions, méthodes, idéologies. Actes des Quatrièmes Journées des Droits Linguistiques(Teramo, Giulianova, Rosciano, Villa Badessa, 20-23 mai 2010)
No full textCe volume réunit des articles traitant des droits linguistiques par rapport à l’enseignement des langues locales, ainsi que d’autres contributions scientifiques sur la même thématique issues d’un colloque international sur « L’enseignement des langues locales : institutions, méthodes, idéologies », qui s’est tenu en Italie, du 20 au 23 mai 2010, sous l’égide de l’association LEM-Italia et de l’Université de Teramo. Par « langues locales » on entend ici tout système linguistique n’ayant pas le statut de langue officielle, donc juridiquement contraignante pour l’État et ses citoyens dans les domaines de l’usage officiel (législation, justice, administration publique, enseignement), ni les mêmes conditions de survie
Dominance sémique, latence du schème et motivation dans les langues régionales: le lexique des outils traditionnels dans le parler de Gussola (Crémone, Italie)
No full textUnsupervised domain adaptation for mobile semantic segmentation based on cycle consistency and feature alignment
Full text linkThe supervised training of deep networks for semantic segmentation requires a huge amount of labeled real world data. To solve this issue, a commonly exploited workaround is to use synthetic data for training, but deep networks show a critical performance drop when analyzing data with slightly different statistical properties with respect to the training set. In this work, we propose a novel Unsupervised Domain Adaptation (UDA) strategy to address the domain shift issue between real world and synthetic representations. An adversarial model, based on the cycle consistency framework, performs the mapping between the synthetic and real domain. The data is then fed to a MobileNet-v2 architecture that performs the semantic segmentation task. An additional couple of discriminators, working at the feature level of the MobileNet-v2, allows to better align the features of the two domain distributions and to further improve the performance. Finally, the consistency of the semantic maps is exploited. After an initial supervised training on synthetic data, the whole UDA architecture is trained end-to-end considering all its components at once. Experimental results show how the proposed strategy is able to obtain impressive performance in adapting a segmentation network trained on synthetic data to real world scenarios. The usage of the lightweight MobileNet-v2 architecture allows its deployment on devices with limited computational resources as the ones employed in autonomous vehicles
- …
