Günter Westphal

Learn More
We present a pattern recognizer to classify a variety of objects and their pose on a table from real world images. Learning of weights in a linear discriminant is based on estimating the relative information contributed by a set of features to the final decision. Evaluation of the discriminant is very fast, allowing for about three decisions per second on(More)
We present an object recognition system built on a combination of feature- and correspondence-based pattern recognizers. The feature-based part, called preselection network, is a single-layer feedforward network weighted with the amount of information contributed by each feature to the decision at hand. For processing arbitrary objects, we employ small,(More)
Acknowledgments I cannot even imagine where I would be today were it not for that handful of friends. the privilege to work in his system biophysics group for almost four years and I fondly enjoyed every single second of it. I immensely enjoyed the superb working conditions, the creative atmosphere, the free flow of ideas, the freedom to develop my own(More)
  • 1