Research and development of methods for the efficient and secure implementation of machine learning methods for machine vision applications. Advanced development and construction of concept architectures for interactive generation of training data and architectures for learning with encrypted data.
Your tasks will cover the conception and design of algorithms for object detection and tracking, including the analysis, implementation and evaluation of the methods. You actively pursue research in the field of video image data analysis.
Typical tasks range from the automated creation of (semantic) 2D and 3D environment models using ground-, drone- or satellite-based imagery to the georegistration of the corresponding data and results. You will develop current state-of-the-art methods from machine learning and computer vision to solve problems in one of the above-mentioned tasks, such as image-based segmentation or image-based 3D reconstruction.
Software implementation and applied research of methods for 3D object recognition and 3D environment modeling based on LiDAR sensor data. Work on an existing mobile multi-sensor system demonstrator (sensor vehicle) for sensor data acquisition and real-time analysis.
In the department Object Recognition we are working on the reconstruction of urban areas using multi-date satellite images. With the rapid advancement of computer vision there are frequently new possibilities available that allow to improve existing parts of our current processing pipeline. Thus, we are offering several master's theses for improving specific sub-problems with current state-of-the-art techniques.
In the department Object Recognition we are working on the reconstruction of urban areas using multi-date satellite images. With the rapid advancement of computer vision there are frequently new possibilities available that allow to improve existing parts of our current processing pipeline. Thus, we are offering several Student Assistant / Hiwi positions for improving specific sub-problems with current state-of-the-art techniques.