TU Berlin

Methods of Geoinformation ScienceDeep Learning for Geographical Data

Page Content

to Navigation

Deep Learning for Geographical Data

Students have a fundamental understanding of the principles, structure, and working of neural networks and their common architectures for solving tasks such as classification, regression, localization, recognition, segmentation, etc., in the context of geoinformation related applications. They are able to plan, implement, train, and evaluate the performance of application-specific deep neural network architectures for processing geographical 2D/3D data such as raster, vector, network, and point cloud data. The students are well acquainted with the established software libraries for Deep Learning and are able to use them self-reliantly.

Deep Learning for Geographical Data (6 CP)
Lectures
Deep Learning for Geographical Data  VL 3633 L 9072 UE 3633 L 9073
Module type
Compulsory elective
Person
in charge
Prof. Dr.-Ing. Martin Kada
Entry
requirements
Computer literacy
Duration
1 semester
Examination
Oral exam
Workload
Overall attendance: 15 x 4 h = 60 h
Preparation and post processing: 15 x 8 h = 120 h

Navigation

Quick Access

Schnellnavigation zur Seite über Nummerneingabe