Now showing items 1-5 of 5

    • An automotive case study on the limits of approximation for object detection 

      Caro Roca, Martí; Tabani, Hamid; Abella Ferrer, Jaume; Moll Echeto, Francisco de Borja; Morancho Llena, Enrique; Canal Corretger, Ramon; Altet Sanahujes, Josep; Calomarde Palomino, Antonio; Cazorla Almeida, Francisco Javier; Rubio Romano, Antonio; Fontova Muste, Pau; Fornt Mas, Jordi (2023-05)
      Article
      Restricted access - publisher's policy
      The accuracy of camera-based object detection (CBOD) built upon deep learning is often evaluated against the real objects in frames only. However, such simplistic evaluation ignores the fact that many unimportant objects ...
    • An energy-efficient GeMM-based convolution accelerator with on-the-fly im2col 

      Fornt Mas, Jordi; Fontova Muste, Pau; Caro Roca, Martí; Abella Ferrer, Jaume; Moll Echeto, Francisco de Borja; Altet Sanahujes, Josep; Studer, Christoph (2023-11)
      Article
      Open Access
      Systolic array architectures have recently emerged as successful accelerators for deep convolutional neural network (CNN) inference. Such architectures can be used to efficiently execute general matrix–matrix multiplications ...
    • An evaluation of energy vs accuracy tradeoffs for object detection accelerators 

      Caro Roca, Martí (Universitat Politècnica de Catalunya, 2022-07-01)
      Master thesis
      Restricted access - author's decision
    • Energy efficient object detection for automotive applications with YOLOv3 and approximate hardware 

      Fornt Mas, Jordi; Fontova Muste, Pau; Caro Roca, Martí; Abella Ferrer, Jaume; Moll Echeto, Francisco de Borja; Altet Sanahujes, Josep; Rubio Sola, Jose Antonio (Institute of Electrical and Electronics Engineers (IEEE), 2023)
      Conference report
      Open Access
      Deep neural networks are the dominant models for perception tasks in the automotive domain, but their high computational complexity makes it difficult to execute them in real time with an acceptable power consumption on ...
    • Unitats funcionals aproximades per a processadors de baix consum 

      Caro Roca, Martí (Universitat Politécnica de Catalunya, 2020-01)
      Bachelor thesis
      Open Access
      Actualment els multiplicadors són una de les unitats funcionals que requereixen més consum d'energia degut al gran nombre de portes lògiques que contenen. Aprofitant el fet que diverses aplicacions poden tolerar un cert ...