Now showing items 1-20 of 1479

    • 2D to 3D body pose estimation for sign language with Deep Learning 

      Perez Granero, Pol (Universitat Politècnica de Catalunya, 2020-06)
      Bachelor thesis
      Open Access
      This project aims at leveraging the challenge of using 3D poses for Sign Language translation or animation by transforming 2D pose datasets into 3D ones. The goal is, using a 3D dataset of American Sign Language, to train ...
    • 3D Bounding box detection from monocular images 

      Catà Villà, Marcel (Universitat Politècnica de Catalunya, 2019-05-08)
      Master thesis
      Open Access
      Object detection is particularly important in robotic applications that require interaction with the environment. Although 2D object detection methods obtain accurate results, these are not enough to provide a complete ...
    • 3D hand reconstruction from RGB-D/RGB video frames in real-time 

      Toda Mas, Aleix (Universitat Politècnica de Catalunya, 2019-04-30)
      Master thesis
      Open Access
      Covenantee:   École polytechnique fédérale de Lausanne
      We present a pipeline able to extract 3D real-world measurements from RGB-D images with high accuracy in real-time. A new method bottom-up multi-person for hand-body pose detection evolved from OpenPifPaf has been presented. ...
    • 3D Human pose, shape and texture from low-resolution images and videos 

      Xu, Xiangyu; Chen, Hao; Moreno-Noguer, Francesc; Jeni, Lázló; De La Torre, Fernando (2022)
      Article
      Open Access
      3D human pose and shape estimation from monocular images has been an active research area in computer vision. Existing deep learning methods for this task rely on high-resolution input, which however, is not always available ...
    • 3D point cloud correspondences using deep learning 

      Rica Palma, Javier de la (Universitat Politècnica de Catalunya, 2018-05)
      Bachelor thesis
      Open Access
      The main goal of the project is to find correspondences between points in two 3D point clouds using deep learning. A deep learning network is trained to select key points and find correspondences between two point clouds ...
    • 3D point cloud correspondences using graph neural networks 

      Gimenez Arnal, Mario (Universitat Politècnica de Catalunya, 2018-10-17)
      Bachelor thesis
      Open Access
      The purpose of this project is the study of neural networks, their training and application together with the creation of a database appropriate to the system. Specically, the 3D convolutional networks applied to a database ...
    • 3D scene reconstruction and understanding from single shot pictures 

      García González, Alfredo (Universitat Politècnica de Catalunya, 2012-09)
      Master thesis
      Open Access
      Augmented reality mixes computer generated graphics with real imaging using computer vision techniques. However, nowadays, augmented reality is still a very young field of research, and its applications usually involve ...
    • 5G-CLARITY: Integrating 5GNR, WiFi and LiFi in private networks with slicing support 

      Camps Mur, Daniel; Ghoraishi, Mir; Gutiérrez Terán, Jesús; Ordoñez-Lucena, Jose; Cogalan, Tezcan; Haas, Harald; García Gómez, Antonio; Sark, Vladica; Aumayr, Erik; van der Meer, Sven; Yan, Shuangyi; Mourad, Alain; Adamuz-Hinojosa, Oscar; Pérez Romero, Jordi; Granda Trigo, Miguel; Bian, Rui (European Conference on Networks and Communications (EuCNC), 2020)
      Conference lecture
      Open Access
      This paper introduces 5G-CLARITY, a 5G-PPP project exploring beyond 5G private networks integrating heterogeneous wireless access including 5GNR, WiFi, and LiFi. The project targets enhancements to current 5GNR performance ...
    • A benchmark for graph neural networks for computer network modeling 

      Carol Bosch, Sergi (Universitat Politècnica de Catalunya, 2019-06-27)
      Master thesis
      Open Access
      Today, network operators still lack functional network models able to make accurate predictions of end-to-end Key Performance Indicators (e.g., delay).This thesis introduces the benchmark for computer network modeling using ...
    • A BF16 FMA is all you need for DNN training 

      Osorio Ríos, John Haiber; Armejach Sanosa, Adrià; Petit, Eric; Henry, Greg; Casas Guix, Marc (Institute of Electrical and Electronics Engineers (IEEE), 2022-07-01)
      Article
      Open Access
      Fused Multiply-Add (FMA) functional units constitute a fundamental hardware component to train Deep Neural Networks (DNNs). Its silicon area grows quadratically with the mantissa bit count of the computer number format, ...
    • A comparative study of demand forecasting models for an online retailer business 

      Lorenzo Villagrasa, Raul (Universitat Politècnica de Catalunya, 2021-06-30)
      Master thesis
      Restricted access - author's decision
    • A comparison between machine learning and classic algorithms for GDP forecast 

      Monar Aguilar, Kenny Xavier (Universitat Politècnica de Catalunya / Universitat de Barcelona, 2022-10)
      Master thesis
      Open Access
      In the recent years there has been an explosive increase in the number of research papers using machine learning methods for forecasting. In this work, I will focus on comparing the estimation of GDP using classical and ...
    • A comprehensive survey of V2X cybersecurity mechanisms and future research paths 

      Sedar, Mohottige Roshan Madhusanka; Kalalas, Charalampos; Vazquez Gallego, Francisco; Alonso Zárate, Luis Gonzaga; Alonso Zarate, Jesús (Institute of Electrical and Electronics Engineers (IEEE), 2023-01-26)
      Article
      Open Access
      Recent advancements in vehicle-to-everything (V2X) communication have notably improved existing transport systems by enabling increased connectivity and driving autonomy levels. The remarkable benefits of V2X connectivity ...
    • A constellation of horrors: analysis and visualization of the #Cuéntalo movement 

      Bucalo Mana, María Soledad; Calvo, Luz; Cucchietti, Fernando; García Povedano, David; García Sáez, Artur; Gómez Celis, Juan Felipe; Arcadio González, Camilo; Marín, Guillermo; Reyes, Patricio; Serra Burriel, Feliu; Vélez García, Diana Fernanda; Meta, Irene (Association for Computing Machinery (ACM), 2019)
      Conference report
      Open Access
      In this work, we analyze content and structure of the Twitter trending topic #cuentalo with the purpose of providing a visualization of the movement. A supervised learning methodology is used to train the classifying ...
    • A Convolutional Neural Network for the automatic diagnosis of collagen VI-related muscular dystrophies 

      Rodríguez Bazaga, Adrián; Roldán Molina, Mónica; Badosa Gallego, Maria del Carmen; Jiménez Mallebrera, Cecilia; Porta Pleite, Josep Maria (2019-12-01)
      Article
      Open Access
      The development of machine learning systems for the diagnosis of rare diseases is challenging, mainly due to the lack of data to study them. This paper surmounts this obstacle and presents the first Computer-Aided Diagnosis ...
    • A cross-layer review of deep learning frameworks to ease their optimization and reuse 

      Tabani, Hamid; Pujol Torramorell, Roger; Abella Ferrer, Jaume; Cazorla Almeida, Francisco Javier (Institute of Electrical and Electronics Engineers (IEEE), 2020)
      Conference report
      Open Access
      Machine learning and especially Deep Learning (DL) approaches are at the heart of many domains, from computer vision and speech processing to predicting trajectories in autonomous driving and data science. Those approaches ...
    • A damage classification approach for structural health monitoring using machine learning 

      Tibaduiza Burgos, Diego Alexander; Torres-Arredondo, Miguel Ángel; Vitola Oyaga, Jaime; Anaya Vejar, Maribel; Pozo Montero, Francesc (2018-12-02)
      Article
      Open Access
      Inspection strategies with guided wave-based approaches give to structural health monitoring (SHM) applications several advantages, among them, the possibility of the use of real data from the structure which enables ...
    • A data-driven wall-shear stress model for LES using gradient boosted decision trees 

      Radhakrishnan, Sarath; Adu Gyamfi, Lawrence; Miró Jané, Arnau; Font García, Bernat; Calafell Sandiumenge, Joan; Lehmkuhl Barba, Oriol (Springer Nature, 2021)
      Conference report
      Open Access
      With the recent advances in machine learning, data-driven strategies could augment wall modeling in large eddy simulation (LES). In this work, a wall model based on gradient boosted decision trees is presented. The model ...
    • A decision making support tool: The resilience management fuzzy controller 

      González Cardenas, Rubén; Nebot Castells, M. Àngela; Múgica Álvarez, Francisco; Vellido Alcacena, Alfredo (Institute of Electrical and Electronics Engineers (IEEE), 2017)
      Conference report
      Open Access
      In this paper a fuzzy controller capable to perform an automated estimation of the period of time necessary to recover a resilience level is proposed. Estimations where made by considering realistic time-dependent action ...
    • A deep learning approach to portfolio optimization 

      Cartanyà Caro, Pau (Universitat Politècnica de Catalunya, 2022-02)
      Bachelor thesis
      Open Access
      Covenantee:   Hong Kong University of Science and Technology
      Des del desenvolupament de la teoria moderna de carteres de Markowitz l'any 1952, s'han dut a terme numerosos avenços per a millorar-ne les tècniques originals, fins al punt que la recerca actual en aquest àmbit es centra ...