Object Recognition with Less Requirements

In recent years remarkable results have been achieved in the field of object recognition. Recognition performance of more than 90% are not rare anymore leading to the conclusion of an application beyond scientific fields. However, such a high performance is often a result of unrealistic constraints of the images to be recognized and the use-cases which are only applicable in controlled laboratory environments. In this paper we propose a system working even under difficult conditions and achieving higher recognition performance as compared to other state-of-the-art systems.