Now showing items 1-20 of 78

  • Action recognition in videos 

    Xu, Zineng (Universitat Politècnica de Catalunya, 2018-06-18)
    Master thesis
    Open Access
    In this project, our work can be divided into two parts: RGB-D based action recognition in trimmed videos and temporal action detection in untrimmed videos.
  • Active garment recognition and target grasping point detection using deep learning 

    Corona Puyane, Enric; Alenyà Ribas, Guillem; Gabas Nova, Antonio; Torras, Carme (2018-02-01)
    Article
    Open Access
    Identification and bi-manual handling of deformable objects, like textiles, is one of the most challenging tasks in the field of industrial and service robotics. Their unpredictable shape and pose makes it very difficult ...
  • A deep analysis on age estimation 

    Huerta Casado, Iván; Fernandez Tena, Carles; Segura, Carlos; Hernando Pericás, Francisco Javier; Prati, Andrea (2015-12-15)
    Article
    Open Access
    The automatic estimation of age from face images is increasingly gaining attention, as it facilitates applications including advanced video surveillance, demographic statistics collection, customer profiling, or search ...
  • Assessing knee OA severity with CNN attention-based end-to-end architectures 

    Górriz, Marc; Antony, Joseph; McGuinness, Kevin; Giró Nieto, Xavier; O'Connor, Noel (2019)
    Conference lecture
    Open Access
    This work proposes a novel end-to-end convolutional neural network (CNN) architecture to automatically quantify the severity of knee osteoarthritis (OA) using X-Ray images, which incorporates trainable attention modules ...
  • Automated curation of brand-related social media images with deep learning 

    Tous Liesa, Rubén; Gómez Parada, Mauro; Poveda, Jonatan; Cruz, Leonel; Wust, Otto; Makni, Mouna; Ayguadé Parra, Eduard (2018-10)
    Article
    Open Access
    This paper presents a work consisting in using deep convolutional neural networks (CNNs) to facilitate the curation of brand-related social media images. The final goal is to facilitate searching and discovering user-generated ...
  • Automatic generation of workload profiles using unsupervised learning pipelines 

    Buchaca Prats, David; Bernal García, Josep Lluis; Carrera Pérez, David (2017-12-27)
    Article
    Open Access
    The complexity of resource usage and power consumption on cloud-based applications makes the understanding of application behavior through expert examination difficult. The difficulty increases when applications are seen ...
  • Automatic speech recognition with deep neural networks for impaired speech 

    España-i-Bonet, Cristina; Rodríguez Fonollosa, José Adrián (Springer, 2016)
    Conference report
    Open Access
    Automatic Speech Recognition has reached almost human performance in some controlled scenarios. However, recognition of impaired speech is a difficult task for two main reasons: data is (i) scarce and (ii) heterogeneous. ...
  • Bioinformatics and medicine in the era of deep learning 

    Bacciu, Davide; Lisboa, Paulo J G; Martín, José David; Stoean, Ruxandra; Vellido Alcacena, Alfredo (I6doc.com, 2018)
    Conference report
    Restricted access - publisher's policy
    Many of the current scientific advances in the life sciences have their origin in the intensive use of data for knowledge discovery. In no area this is so clear as in bioinformatics, led by technological breakthroughs in ...
  • Blood pressure assessment with differential pulse transit time and deep learning: a proof of concept 

    Ribas Ripoll, Vicent; Vellido Alcacena, Alfredo (Karger, 2019-02)
    Article
    Open Access
    Modern clinical environments are laden with technology devices continuously gathering physiological data from patients. This is especially true in critical care environments, where life-saving decisions may have to be made ...
  • Clasificación de imágenes histológicas mediante redes neuronales convolucionales 

    Bustos Pelegrí, Joel (Universitat Politècnica de Catalunya, 2018-05)
    Bachelor thesis
    Open Access
    Nowadays, one of the most important goals in the field of medicine, is to be able to develop some kind of system that helps to automate the classification and detection of different cancer pathologies in tissue images. ...
  • Coarse grain parallelization of deep neural networks 

    González Tallada, Marc (Institute of Electrical and Electronics Engineers (IEEE), 2016)
    Conference lecture
    Restricted access - publisher's policy
    Deep neural networks (DNN) have recently achieved extraordinary results in domains like computer vision and speech recognition. An essential element for this success has been the introduction of high performance computing ...
  • Comparing fixed and adaptive computation time for recurrent neural networks 

    Fojo, Daniel; Campos Camunez, Victor; Giró Nieto, Xavier (2018)
    Conference report
    Open Access
    Deep networks commonly perform better than shallow ones, but allocating the proper amount of computation for each particular input sample remains an open problem. This issue is particularly challenging in sequential tasks, ...
  • Dancing with deep learning 

    Arnal Romero, Gisela (Universitat Politècnica de Catalunya, 2017-06)
    Bachelor thesis
    Open Access
    Art and technology are in constant synergetic expansion. Willy Barleycorn - Machine Learning in Dance. When art uses technology as a way of creation by the hand of the own artist it comes a synergy with great potential. ...
  • Deep and cognitive learning applied to Precision Medicine: the initial experiments linking (epi)genome to phenotypes-disease characteristics 

    Cirillo, Davide; Valencia, Alfonso (Barcelona Supercomputing Center, 2017-05-04)
    Conference report
    Open Access
    Present-day era of Big Data provides the unique opportunity to develop innovative approaches for data analysis to find new insights into specialized fields of biomedical research such as Precision Medicine [1]. Precision ...
  • Deep learning backend for single and multisession i-vector speaker recognition 

    Ghahabi, Omid; Hernando Pericás, Francisco Javier (2017-04-01)
    Article
    Open Access
    The lack of labeled background data makes a big performance gap between cosine and Probabilistic Linear Discriminant Analysis (PLDA) scoring baseline techniques for i-vectors in speaker recognition. Although there are some ...
  • Deep learning for detecting freezing of gait episodes in Parkinson’s disease based on accelerometers 

    Camps, Julià; Samà Monsonís, Albert; Martín Muñoz, Mario; Rodríguez Martín, Daniel Manuel; Pérez López, Carlos; Alcaine, Sheila; Mestre, Berta; Prats, Anna; Crespo, M. Cruz; Cabestany Moncusí, Joan; Bayés, Àngels; Català Mallofré, Andreu (2017)
    Conference report
    Open Access
    Freezing of gait (FOG) is one of the most incapacitating symptoms among the motor alterations of Parkinson’s disease (PD). Manifesting FOG episodes reduce patients’ quality of life and their autonomy to perform daily living ...
  • Deep learning for freezing of gait detection in Parkinson’s disease patients in their homes using a waist-worn inertial measurement unit 

    Camps, Julià; Samà Monsonís, Albert; Martín Muñoz, Mario; Rodríguez Martín, Daniel Manuel; Pérez López, Carlos; Moreno Aróstegui, Juan Manuel; Cabestany Moncusí, Joan; Català Mallofré, Andreu; Alcaine, Sheila; Mestre, Berta; Prats, Anna; Crespo, M. Cruz; Counihan, Timothy; Browne, Patrick; Quinlan, Leo R.; ÓLaighin, Gearóid; Sweeney, Dean; Lewy, Hadas; Vainstein, Gabriel; Costa, Alberto; Annicchiarico, Roberta; Bayés, Àngels; Rodríguez Molinero, Alejandro (2017-10-16)
    Article
    Restricted access - publisher's policy
    Among Parkinson’s disease (PD) motor symptoms, freezing of gait (FOG) may be the most incapacitating. FOG episodes may result in falls and reduce patients’ quality of life. Accurate assessment of FOG would provide objective ...
  • Deep learning for multimedia processing-Predicting media interestingness 

    Cardoner Campi, Lluc (Universitat Politècnica de Catalunya, 2017-06-30)
    Bachelor thesis
    Open Access
    Covenantee:  Technische Universität Wien
    This thesis explores the application of a deep learning approach for the prediction of media interestingness. Two different models are investigated, one for the prediction of image and one for the prediction of video ...
  • Deep Learning for semantic segmentation of airplane hyperspectral imaging 

    Balibrea Rull, Mar (Universitat Politècnica de Catalunya, 2019-06)
    Bachelor thesis
    Open Access
    Given their success, both qualitative and quantitative, Deep Neural Networks have been used to approach classification and segmentation problems for images, especially during these last few years where it has been possible ...
  • Deep learning for ultrasound data-rate reduction 

    Sainz Lorenzo, Yeray (Universitat Politècnica de Catalunya, 2019-02-19)
    Master thesis
    Open Access
    US devices generate a set of signals that are carried from a transducer probe to a computer for further processing in order to obtain images. Those signals are transmitted between both ends through a set of cables, making ...