Logo des Repositoriums
 

Künstliche Intelligenz 29(1) - März 2015

Autor*innen mit den meisten Dokumenten  

Auflistung nach:

Neueste Veröffentlichungen

1 - 10 von 13
  • Zeitschriftenartikel
    News
    (KI - Künstliche Intelligenz: Vol. 29, No. 1, 2015)
  • Zeitschriftenartikel
    Attentional Scene-Exploration and Object Discovery in Image and RGB-D Data
    (KI - Künstliche Intelligenz: Vol. 29, No. 1, 2015) Martín García, Germán; Werner, Thomas; Frintrop, Simone
    In this paper, we summarize our project work of the last two years, where we addressed the tasks of visually exploring a scene with visual attention mechanisms based on saliency computation, and of locating unknown objects in the environment. The latter is also called object discovery and consists in finding candidate objects without previous knowledge about the objects themselves or the scene. We follow an approach motivated from human perception and combine saliency and segmentation to generate object candidates. We show results on 2D images as well as on 3D sequences obtained from an RGB-D camera.
  • Zeitschriftenartikel
    Beyond Simple and Complex Neurons: Towards Intermediate-level Representations of Shapes and Objects
    (KI - Künstliche Intelligenz: Vol. 29, No. 1, 2015) Rodríguez-Sánchez, Antonio; Neumann, Heiko; Piater, Justus
    Knowledge of the brain has much advanced since the concept of the neuron doctrine developed by Ramón y Cajal (R Trim Histol Norm Patol 1:33–49, 1888). Over the last six decades a wide range of functionalities of neurons in the visual cortex have been identified. These neurons can be hierarchically organized into areas since neurons cluster according to structural properties and related function. The neurons in such areas can be characterized to a first order approximation by their (static) receptive field function, viz their filter characteristic implemented by their connection weights to neighboring cells. This paper aims to provide insights on the steps that computer models in our opinion must pursue in order to develop robust recognition mechanisms that mimic biological processing capabilities beyond the level of cells with classical simple and complex receptive field response properties. We stress the importance of intermediate-level representations to achieve higher-level object abstraction in the context of feature representations, and summarize two current approaches that we consider are advances toward achieving that goal.
  • Zeitschriftenartikel
    Towards an Embodied Developing Vision System
    (KI - Künstliche Intelligenz: Vol. 29, No. 1, 2015) Atıl, İlkay; Kalkan, Sinan
    Many cognitive scientists now agree that artificial cognition might be probably achieved developmentally, starting from a set of basic-level premature capabilities and incrementally self-extending itself with experience through discrete or continuous stages bred with experience. Although we are still far from seeing an artificial full-fledged self-extending cognitive system, the literature has provided promising examples and demonstrations. Nonetheless, not much thought is given to the modeling of how an artificial vision system, an important part of a developing cognitive system, can develop itself in a similar manner. In this article, we dwell upon the issue of a developing vision system, the relevant problems and possible solutions whenever possible.
  • Zeitschriftenartikel
    Technische und rechtliche Betrachtungen zur Autonomie kooperativ-intelligenter Softwareagenten
    (KI - Künstliche Intelligenz: Vol. 29, No. 1, 2015) Kirn, Stefan; Müller-Hengstenberg, Claus D.
    Die Automatisierung technischer Systeme basiert auf präzisen Definitionen für deren Verhalten und der ebenso präzisen Umsetzung in zuverlässige und robuste Lösungen. Zukünftige Automatisierungsansätze, wie bspw. das autonome Fahren [vgl. Bundesanstalt für Straßenbau (2012)], werden jedoch viel umfassender als bisher auf Sensordaten, Kontextwissen, Adaptions- und Interaktionsfähigkeit technischer Systeme angewiesen sein. So befahren “teilautomatisierte” Fahrzeuge im Probebetrieb schon heute ohne Fahrereingriffe das öffentliche Straßennetz. KI-basierte Lösungsansätze spielen dabei eine zunehmend wichtigere Rolle (vgl. u.a. die Ansätze und Ergebnisse der Forschungsprogramme Autonomik und Industrie 4.0), insbesondere eröffnen sie die Möglichkeit, einen technisch geprägten Autonomie-Begriff zu entwickeln, der über das Konzept der Automatisierung hinausreicht. Dies ist allerdings mit zwei Problemen verbunden: Einerseits garantieren nicht alle KI-Methoden Determinismus und Determiniertheit, andererseits kann es bei Interaktion autonomer Systeme (im Folgenden als Agenten bezeichnet) mit ihrer Umwelt selbst dann zu unerwünschten Situationen kommen, wenn diese technisch einwandfrei funktionieren – es sei denn, es gelänge dem Entwickler, alle möglichen Zustände und Verhaltensweisen der Umwelt in vollem Umfang zu antizipieren und im Gesamtsystemmodell abzubilden. Dies wirft rechtliche Fragen auf. Tritt der Mensch damit zunehmend seine Verantwortung für notwendige Entscheidungen in seinem gesellschaftlichen Umfeld an Softwareagenten ab? Kann der Agent für sein Handeln verantwortlich gemacht werden, wird er für die Folgen seines Verhaltens haften? Aber auch, wenn mehrere Agenten bspw. gemeinsam ein Musikstück komponieren: wem gehören die Rechte, müssen für dessen Abspielen ebenfalls GEMA-Gebühren bezahlt werden – und wer bekommt die Einnahmen? Und wer haftet, wenn ein autonomes Fahrzeug (ohne Fahrer) Unfälle verursacht?
  • Zeitschriftenartikel
    Efficient Learning of Pre-attentive Steering in a Driving School Framework
    (KI - Künstliche Intelligenz: Vol. 29, No. 1, 2015) Rudzits, Reinis; Pugeault, Nicolas
    Autonomous driving is an extremely challenging problem and existing driverless cars use non-visual sensing to palliate the limitations of machine vision approaches. This paper presents a driving school framework for learning incrementally a fast and robust steering behaviour from visual gist only. The framework is based on an autonomous steering program interfacing in real time with a racing simulator: hence the teacher is a racing program having perfect insight into its position on the road, whereas the student learns to steer from visual gist only. Experiments show that (i) such a framework allows the visual driver to drive around the track successfully after a few iterations, demonstrating that visual gist is sufficient input to drive the car successfully; and (ii) the number of training rounds required to drive around a track reduces when the student has experienced other tracks, showing that the learnt model generalises well to unseen tracks.
  • Zeitschriftenartikel
    What We Can Learn From the Primate’s Visual System
    (KI - Künstliche Intelligenz: Vol. 29, No. 1, 2015) Krüger, Norbert; Zillich, Michael; Janssen, Peter; Buch, Anders Glent
    In this review, we discuss the impact (or lack thereof) biologically motivated vision has had on computer vision in the last decades. We then summarize a number of computer vision and robotic problems for which biological models can give indications for how these can be addressed. Then we summarize important findings about the primate’s visual system and draw a number of conclusions for the development of algorithms from these findings.
  • Zeitschriftenartikel
    Special Issue on Bio-inspired Vision Systems
    (KI - Künstliche Intelligenz: Vol. 29, No. 1, 2015) Zillich, Michael; Krüger, Norbert
  • Zeitschriftenartikel
    Deep Representation Hierarchies for 3D Active Vision
    (KI - Künstliche Intelligenz: Vol. 29, No. 1, 2015) Sabatini, Silvio P.
    Starting from the acknowledged properties of visual cortical neurons, we show how diversified and composite visual descriptors come up from different hierarchical combinations of the harmonic content of the visual signal. The resulting deep hierarchy networks can specialize to solve different tasks and trigger different behaviors, without necessarily getting through an explicit measure of the re-constructive visual attributes of the observed scene. Distinct specializations for stereopis and for active control of the vergence movements of a binocular system are presented. In particular, the advantage of not abandoning distributed representations of multiple solutions to prematurely construct integrated description of cognitive entities and commit the system to a particular behavior is discussed. Pilot CPU-GPU implementations of the proposed cortical-like architectures prove to be promising solutions for the next-generation of robot vision systems, which should be capable of calibrating and adapting autonomously through the interaction with the environment.
  • Zeitschriftenartikel
    A Required Paradigm Shift in Today’s Vision Research
    (KI - Künstliche Intelligenz: Vol. 29, No. 1, 2015) Krüger, Norbert