Projektbeschreibung
Maschinenlerndaten und -algorithmen für maschinelles Sehen neu überdenken
Selbstfahrende Autos scheinen für uns bereits in greifbare Nähe gerückt zu sein, was zum Teil auf den Erfolg von Algorithmen für maschinelles Sehen zurückzuführen ist, die als „Augen“ für derartige autonome Fahrzeuge entwickelt worden sind. Um sicher durch die Welt zu navigieren, müssen autonome Fahrzeuge die dynamischen Objekte in der Umgebung verstehen, d. h. mannigfaltige, sich bewegende Objekte erkennen, segmentieren und nachverfolgen können. Mit maschinellem Sehen kann dieses Problem nun vor allem dank der Fortschritte im Bereich des Deep Learning gelöst werden. Die meisten Verfahren beruhen auf faltenden neuronalen Netzwerken, die auf großen Datensätzen überwacht trainiert werden. Aber reicht dieses Paradigma aus, um die Komplexität unserer Straßen wiederzugeben? Das ERC-finanzierte Projekt DynAI wird über das überwachte Lernen hinausgehen. Die Projektforschenden werden innovative Modelle für maschinelles Lernen entwerfen, die direkt aus unmarkierten Videoströmen lernen können.
Ziel
Computer vision has become a powerful technology, able to bring applications such as autonomous vehicles and social robots closer to reality. In order for autonomous vehicles to safely navigate a scene, they need to understand the dynamic objects around it. In other words, we need computer vision algorithms to perform dynamic scene understanding (DSU), i.e. detection, segmentation, and tracking of multiple moving objects in a scene. This is an essential feature for higher-level tasks such as action recognition or decision making for autonomous vehicles. Much of the success of computer vision models for DSU has been driven by the rise of deep learning, in particular, convolutional neural networks trained on large-scale datasets in a supervised way. But the closed-world created by our datasets is not an accurate representation of the real world. If our methods only work on annotated object classes, what happens if a new object appears in front of an autonomous vehicle? We propose to rethink the deep learning models we use, the way we obtain data annotations, as well as the generalization of our models to previously unseen object classes. To bring all the power of computer vision algorithms for DSU to the open-world, we will focus on three lines of research: 1-Models. We will design novel machine learning models to address the shortcomings of convolutional neural networks. A hierarchical (from pixels to objects) image-dependent representation will allow us to capture spatio-temporal dependencies at all levels of the hierarchy. 2-Data. To train our models, we will create a new large-scale DSU synthetic dataset, and propose novel methods to mitigate the annotation costs for video data. 3-Open-World. To bring DSU to the open-world, we will design methods that learn directly from unlabeled video streams. Our models will be able to detect, segment, retrieve, and track dynamic objects coming from classes never previously observed during the training of our models.
Wissenschaftliches Gebiet
- engineering and technologymechanical engineeringvehicle engineeringautomotive engineeringautonomous vehicles
- natural sciencescomputer and information sciencesartificial intelligencecomputer vision
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringroboticsautonomous robots
- natural sciencescomputer and information sciencesartificial intelligencemachine learningdeep learning
- natural sciencescomputer and information sciencesartificial intelligencecomputational intelligence
Programm/Programme
- HORIZON.1.1 - European Research Council (ERC) Main Programme
Thema/Themen
Finanzierungsplan
ERC - Support for frontier research (ERC)Gastgebende Einrichtung
20124 Milano
Italien