Suchergebnisse

Micromobility is an emerging way of travelling which has revolutionized the transport sector. Its agility and simplicity come from the hand of a high vulnerability to traffic accidents, which requires proper treatment. This thesis proposes the use of state-of-the-art computer vision architectures with multitasking techniques to optimize a lightweight module capable of performing three solutions in real time. Thus, it is proposed to create a system capable of recognizing the driver?s environment, detecting possible hazards on the road and monitoring their position over time. In addition, the module is also capable of classifying the type of way on which the vehicle circulates to adapt vehicle?s speeds to the pertinent legislation. The implementation of the algorithms developed in a mobile application would allow to control the use of micro vehicles, increase safety and stop the current growth of accidents of this new mobility. ; La micromobilitat és una forma emergent de desplaçament que està revolucionant el sector del transport. La seva agilitat i simplicitat vénen de la mà d?una alta vulnerabilitat a accidents de trànsit, que necessita del seu tractament adequat. Aquesta tesi proposa una arquitectura de visió per ordinador que usa de tècniques de multitasking per optimitzar un mòdul lleuger capaç de realitzar tres solucions a temps real. Així, es proposa realitzar un sistema capaç de reconèixer l?entorn del conductor, detectant possibles obstacles a la calçada i monitoritzant la seva posició en el temps. A més, el mòdul és capaç de classificar la tipologia de via en què circula el vehicle per a adaptar la velocitat del vehicle a la legislació pertinent. La implementació dels algoritmes desenvolupats en una aplicació mòbil permetria controlar l?ús de micro vehicles, incrementar la seguretat i aturar aquest creixement actual d?accidents d?aquesta nova mobilitat. ; La micromovilidad es una forma emergente de desplazamiento que está revolucionando el sector del transporte. Su agilidad y simpleza vienen de la mano de una alta vulnerabilidad en accidentes de tráfico, que necesita de su tratamiento adecuado. Esta tesis propone el uso de una arquitectura multitask de visión por computadora para optimizar un módulo ligero capaz de realizar tres soluciones a tiempo real. Así, se propone realizar un sistema capaz de reconocer el entorno del conductor, detectando posibles obstáculos en la calzada y monitorizando su posición en el tiempo. Además, el módulo es capaz de clasificar la tipología de vía en que circula el vehículo para adaptar las velocidades del vehículo a la legislación pertinente. La implementación de los algoritmos desarrollados en una aplicación móvil permitiría controlar el uso de micro vehículos, incrementar la seguridad y parar este crecimiento actual de accidentes de esta nueva movilidad.

High-resolution and variable-shape images have not yet been properly addressed by the AI community. The approach of down-sampling data often used with convolutional neural networks is sub-optimal for many tasks, and has too many drawbacks to be considered a sustainable alternative. In sight of the increasing importance of problems that can benefit from exploiting high-resolution (HR) and variable-shape, and with the goal of promoting research in that direction, we introduce a new family of datasets (MetH). The four proposed problems include two image classification, one image regression and one super resolution task. Each of these datasets contains thousands of art pieces captured by HR and variable-shape images, labeled by experts at the Metropolitan Museum of Art. We perform an analysis, which shows how the proposed tasks go well beyond current public alternatives in both pixel size and aspect ratio variance. At the same time, the performance obtained by popular architectures on these tasks shows that there is ample room for improvement. To wrap up the relevance of the contribution we review the fields, both in AI and high-performance computing, that could benefit from the proposed challenges. ; This work is partially supported by the Intel-BSC Exascale Lab agreement, by the Spanish Government through Programa Severo Ochoa (SEV-2015-0493), by the Spanish Ministry of Science and Technology through TIN2015-65316-P project, and by the Generalitat de Catalunya (contracts 2017-SGR-1414). ; Preprint

This work pretends to find a systematic proximity between Nicholas from Cues & German idealism, particularly, in the notion of "visio intellectualis," understood in the theoretical ground of knowledge. It pretends to demonstrate that what Cusanus thinks as "intellectual vision" ends up being not other thing than the "self-contemplating I" ["selbst anschauende Ich"] in which Schelling thinks and, therefore, that what is at the base of Cusanus' thought & was thought on his time in his way is the same thing that was shown at Schelling's time in Schelling's way. Adapted from the source document.