Neural Network Application

The design motivation is what distinguishes neural networks from other mathematical techniques: A neural network is a processing tool, either an algorithm, or actual hardware, whose design was motivated by the design & functioning of human brains & components thereof.

An Artificial Neural Network is a network of lots of simple processors ("units"), each possibly having a (little amount of) local memory. The units are connected by unidirectional communication channels ("connections"), which carryover numeric (as opposed to symbolic) information. The units operate only on their local information & on the inputs they get by the connections.


There's lots of different types of Neural Networks, each of which has different strengths particular to their applications. The abilities of different networks can be related to their structure, dynamics & learning methods.

Neural Networks offer improved performance over conventional technologies in areas which includes: Machine Vision, Robust Pattern Detection, Signal Filtering, Virtual Reality, Information Segmentation, Information Compression, Information Mining, Text Mining, Artificial Life, Adaptive Control, Optimization & Scheduling, Complex Mapping & more.

Speechreading using neural networks

As part of the research program Neuroinformatik the IPVR develops a neural speechreading method as part of a user interface for a workstation. The two main parts of the method include a face tracker (done by Marco Sommerau), lip modeling & speech processing (done by Michael Vogt) & the development & application of SNNS for neural network training (done by Günter Mamier).

Automatic speechreading is based on a robust lip picture analysis. In this approach, no special illumination or lip make-up is used. The analysis is based on true color video images. The method allows for realtime tracking & storage of the lip region & robust off-line lip model matching. The proposed model is based on cubic outline curves. A neural classifier detects visibility of teeth edges & other attributes. At this stage of the approach the edge between the closed lips is automatically modeled if applicable, based on a neural network's decision.

To accomplish high flexibility during lip-model development, a model description language has been defined & implemented. The language allows the definition of edge models (in general) based on knots & edge functions. Inner model forces stabilize the overall model shape. User defined picture processing functions may be applied along the model edges. These functions & the inner forces contribute to an overall energy function. Adaptation of the model is done by gradient descent or simulated annealing like algorithms. The figure shows four configuration of the lip model, consisting of an upper lip edge as well as a lower lip edge. The model edges are defined by Bezier-functions. Outer control knots stabilize the position of the corners of the mouth.