DLR Portal
Home|Sitemap|Kontakt|Barrierefreiheit Impressum und Nutzungsbedingungen Datenschutz Cookies & Tracking |English
Sie sind hier: Home:IMF:Abteilung: EO Data Science:Team: Visual Learning and Reasoning
Erweiterte Suche
Earth Observation Center
DFD
IMF
Abteilung: Atmosphären-Prozessoren
Abteilung: EO Data Science
Team: Big Data Analytics
Team: Visual Learning and Reasoning
Abteilung: Experimentelle Verfahren
Abteilung: Photogrammetrie und Bildanalyse
Abteilung: SAR-Signalverarbeitung
Services
Technologie
Anwendungen und Projekte
Satellitendaten
Mediathek
News / Archiv
Blogs + Features
Drucken

Team: Visual Learning and Reasoning



The team works on developing vision algorithms for earth observation data that go beyond perception. In particular, this comprises methods like combining image and text information, exploiting the spatial relations of objects, and drawing conclusions from temporal developments and trends.
Currently, the team is conducting research in the following areas:

Vision-Language Reasoning


Most current AI models for earth observation are trained to perform a single task, like classifying a scene into a pre-defined number of classes or detecting objects of pre-defined classes in an image. In order to address this limitation, the team is developing models that can answer natural language queries. It is our vision that this approach will allow non-experts to carry out AI-based analyses on earth observation data by communicating with the models through simple text prompts.

Spatial Reasoning


By exploiting the spatial relationships between objects and areas in earth observation imagery, it is possible to greatly reduce the number of labels needed for the training of neural network models. Conventional deep learning models for the segmentation of remote sensing imagery require a ground truth label for every single pixel in the training dataset. Through the use of spatial reasoning strategies, it is possible to greatly reduce the labelling effort needed, so that the annotator can simply draw single dots, scribbles or polygons instead of annotating all pixels (see figure).

Temporal Reasoning


Sometimes, not only spatial relationships matter, but also how things evolve over time. In temporal reasoning, these connections are taken into account.


Kontakt
Dr. Lichao Mou
Deutsches Zentrum für Luft- und Raumfahrt e.V. (DLR)

Institut für Methodik der Fernerkundung
, EO Data Science
Weßling

Tel.: +49 8153 28-4510

Copyright © 2023 Deutsches Zentrum für Luft- und Raumfahrt e.V. (DLR). Alle Rechte vorbehalten.