J.C.A. van der Lubbe

Facial feature localization and adaptation of a generic face model for model-based coding

A method for the adaptation of a generic 3-D face model to an actual face in a head-and-shoulders scene is discussed, with application to video-telephony. The adaptation is carried out both on a global scale to reposition and resize the wire-frame, as well as on a local scale to mimic individual physiognomy. To this effect a hierarchical scheme is developed to extract the semantic features in the head-and-shoulders scene, such as silhouette, face, eyes and mouth, using a knowledge-based selection mechanism. These algorithms, which are to be an integral part of a general model-based image coder, are tested on typical videophone sequences.

The R-norm information measure

The R -norm information measure is discussed and its properties, as well as an axiomatic characterization, are given. The measure is extended to conditional and joint measures. Applications to coding and hypothesis testing are given. The R -norm information measure includes Shannons information measure as a special case.

Some aspects of error bounds in feature selection

Abstract In this paper we discuss various bounds on the Bayesian probability of error, which are used for feature selection, and are based on distance measures and information measures. We show that they are basically of two types. One type can be related to the f -divergence, the other can be related to information measures. This also clarifies some properties of these measures for the two-class problem and for the multiclass problem. We give some general bounds on the Bayesian probability of error and discuss various aspects of the different approaches.

Transformation of a general 3D facial model to an actual scene face

Accurate modeling and adaptation of scene objects play a crucial role in model-based coding schemes. In this paper a method is presented which transforms a general model of a face to the actual face viewed in the scene. The transformation is based on the facial contours (face border, eyes and mouth) and consists of an affine transformation (for all vertices) and local transformations for each individual vertex. Local transformations are executed taking into consideration the elastic properties of the model.<<ETX>>

Privacy in mobile agents

Agent technology gives many promises for future IT-systems, but little privacy related problems in this technology have been resolved. This paper addresses the privacy problems in agent technology and provides several solutions to overcome some of these problems. It is shown that if an encryption algorithm is used in a different form, it can also provide privacy in communication for agents. Second, it is shown how secret computations can be performed when the function to be executed is a polynomial. Finally, an agent digital signature is described that takes into consideration that the agents private key must be seen as the most privacy sensitive information of an agent.

Information theoretical approach to mobile code

Mobile code is software code that may be run at a remote location, which is possibly untrustworthy. This paper provides a secrecy model for mobile code based on Shannons information theoretic model. Due to different characteristics of mobile code, such as the possibility of plaintext-attacks and untrustworthy execution environments, a new definition for the maximum level of secrecy is necessary. This results in different minimum requirements to obtain this level of secrecy. Furthermore, the unicity distance is derived in various situations

Bivariate certainty and information measures

Abstract The generalized theory of marginal certainty and information measures, as introduced by Van der Lubbe et al. [31], is extended to the conditional and joint cases. Bivariate information measures are introduced with the help of the conditional and joint certainty measures, analogously to the manner in which marginal information measures are derived from marginal certainty measures. This approach leads to new conditional and joint information measures for the well-known marginal measures. Furthermore, this approach unifies the already known bivariate measures with the marginal ones in one generalized probabilistic theory of information measures.

Hierarchical classification inference for fuzzy data analysis

One of the main problems in fuzzy data analysis is the clustering of data. In this paper an expert system approach is followed. On the basis of training data sets a hierarchical knowledge tree is generated consisting of rules that are characterized by an increasing specificity. The hierarchical knowledge is used for inferring decisions on new data sets to be assessed. In order to reduce further the computational complexity the core zone index is introduced, which guarantees the optimal search level in the hierarchical knowledge tree.

Two-layer coding for videotelephone

Abstract Design of a two-layer coding scheme is considered under the constraint that the first layer running at 64 kbit/s is to perform the normal H.261 functions. The second layer is recruited optionally to enhanced the image quality and/or to increase the frame rate. In the chosen scheme, the second laeyr processes the difference between the original signal and the reconstruction of the first layer, by a combination of DCT and pixel correction techniques. While SNR improvements achieved were of the order of 1 dB, subjective tests indicated a much stronger preference for the two-layer codec.