Browsing by Contributor "Dimiccoli, Mariella"

Now showing items 1-6 of 6

    • Estudio e implementación de un algoritmo de image inpainting: priority belief propagation 

      Coll Mulet, Tomeu (Universitat Politècnica de Catalunya, 2009-09)
      Master thesis (pre-Bologna period)
      Open Access
      En este proyecto se presenta un algoritmo que permite realizar tanto image inpainting como síntesis de texturas. Image inpainting consiste básicamente en recuperar de una forma visualmente plausible, una parte o región ...

    • First-person activity recognition: how to generalize to unseen users? 

      Mollova, Emanuela (Universitat Politècnica de Catalunya, 2017)
      Master thesis
      Open Access
      Recent advances in wearable technology, accompanied by the decreasing cost of data storage and increase of data availability have made possible to take pictures everywhere at every time. Wearable cameras are nowadays ...

    • Human motion dataset in the wild (MAT) 

      Coll Ribes, Gabriel (Universitat Politècnica de Catalunya, 2019-03-07)
      Bachelor thesis
      Open Access
      The popularity of wearable cameras is steadily increasing, both for entertainment and productivity purposes. Understanding the footage and inferring the body pose of the camera wearer can be of great importance in many ...

    • Improving object detection by exploiting semantic relations between objects 

      Petrov, Yordan (Universitat Politècnica de Catalunya, 2017-05)
      Master thesis
      Open Access
      Object detection is a fundamental and challenging problem in computer vision. Detecting the objects visible in an image can give us a good understanding and description of the image. The extracted information can later ...

    • Learning graph-based event representations for unconstrained video segmentation 

      Tura Vecino, Biel (Universitat Politècnica de Catalunya, 2021-05)
      Master thesis
      Restricted access - confidentiality agreement
      Covenantee:   Institut de Robòtica i Informàtica Industrial
      Recent research has shown that, in particular domains, unsupervised learning algorithms are achieving on par, or even better performance than fully supervised algorithms, avoiding the need of human labelled data. The ...

    • Social relation recognition in egocentric photo-strems 

      Sanchez Aimar, Emanuel (Universitat Politècnica de Catalunya, 2018-04-17)
      Master thesis
      Open Access
      Covenantee:   Universitat de Barcelona / Universitat Rovira i Virgili
      Presentamos un método para la categorización automática de relaciones sociales en secuencias de imágenes egocéntricas. Por cada imagen, extraemos características visuales con redes convolucionales (atributos como edad, ...