Research Interests
Scene Interpretation
- Probabilistic Inference
- Combining Interpretation with Classification
Other
Projects
eTraining for Interpreting Images of Man-Made Scenes (link)
The aim of this project is to advance the state of the art of cognitive systems by developing a methodology for autonomous and continuous learning. The project will concentrate on structural learning, where relations between components and compositional hierarchies play a central role in object categorization. Such learning is particularly relevant for the interpretation of man-made objects, hence the project will use the recognition of buildings in outdoor scenes as its exemplary application domain.
Involvement in the project: Apr 2008 - Oct 2009
Publications
Arne Kreutzmann; Kasim Terzić; Bernd Neumann: Context-aware Classification for Incremental Scene Interpretation, Workshop on Use of Context in Vision Processing (UCVP 2009), Boston, Nov 2009 [bibtex][PDF]
Movies
Context-aware Classification in Action (link)
The video demonstartes the algorithm that was developed in Context-aware Classification for Incremental Scene Interpretation (see Publications). First all the detections as found by a stable region detector as well as one for interesting regions are shown and then every piece is classified or rejected as a detection not corresponding to an object (False-Positive).