Seguí Mesquida, SantiDomínguez Mira, Ricardo2019-06-212019-06-212019-01-18https://hdl.handle.net/2445/135720Treballs Finals de Grau de Matemàtiques, Facultat de Matemàtiques, Universitat de Barcelona, Any: 2019, Director: Santi Seguí Mesquida[en] Generative Adversarial Networks (GANs) have been widely used for image generation. From the conception of this model in 2014 until now, there has been a increasingly interest in this model, which have lead it to be one of the most popular Deep Learning models nowadays. The popularity of this model has entailed a significant list of variations of it. This work aims are to study the GAN model, and specifically, one of its variations: TripletGAN.59 p.application/pdfengcc-by-nc-nd (c) Ricardo Domínguez Mira, 2019http://creativecommons.org/licenses/by-nc-nd/3.0/es/Aprenentatge automàticTreballs de fi de grauResolució de problemesSistemes d'imatgesAlgorismes computacionalsMachine learningBachelor's thesesProblem solvingImaging systemsComputer algorithmsinfo:eu-repo/semantics/bachelorThesisinfo:eu-repo/semantics/openAccess