The previously-discussed project scope including holding a technology study, the development of a research prototype and evaluation and action. Background of the research project is the idea that the necessary measurements for the Gehassistenten can be provided by a single-mounted shoe camera (single camera system). The advantage over a previously envisaged 3D camera system would be in the smaller design. The required data (recognition of curbs, detection of obstacles, etc.) is expected to generate 2D image analysis methods. Based on a literature review (see Annex) have advanced Deep Learning Methods with convolutional neural network (CNN) operate the corresponding potential for it.
|Effective start/end date||15/08/16 → 15/04/17|
Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.