A. Geurtz

Coding of digital TV by motion-compensated Gabor decomposition

The problem of easily implementable filter bank design for subband coding is discussed. A technique is proposed to generate filter banks with coefficients in powers-of-two and with maximum localization. The hierarchical structure of the filter bank is then used to eliminate the temporal redundancies by using a multiresolution motion estimation technique. Simulations show a robust estimation of the motion field with a relatively simple implementation.

Efficient image reconstruction from sign and energy information

In recent year several algorithms for signal reconstruction from sign information have been developed. In this paper we propose an improved version of the 1D algorithm presented, with the following new characteristics: (1) restriction of the input data space to the set of energy normalized bandpass functions, (2) preservation of the global energy content of the original signal, and (3) implicit determination of the scaling constant. Simulation results show that the restriction of the solution data space to the set on energy normalized functions leads, in general, to a reconstruction that, even if is not coincident in all the zerocrossing positions to the original signal, seems to be stable in L2 sense. A complete coding scheme based on the JBIG standard for an efficient encoding of the zerocrossing is presented. Encouraging results have been obtained both at reconstruction and at coding level that show the feasibility of the global scheme.

Image coding based on zerocrossing and energy information

In this paper we propose a complete subband-based codec for image compression purposes, using a new 2D algorithm based on zero-crossing and energy information. Although signal representation based on sign information is attractive from a coding point of view, two major problems inhibited its practical use, namely the large amount of side information needed for obtaining a stable reconstruction and the efficient treatment of inherently 2D signals. The new algorithm significantly improves the reconstruction results obtained in our previous work, thanks to its improved convergence and its capabilities to deal with the two-dimensional nature of the subband image signals. The zero-crossing information is compressed in a lossless way using the JBIG standard and the energy information in a lossy way using an arithmetic coder. Good results have been obtained both at the reconstruction and at the coding level, which show the feasibility of the global scheme.

Use of wavelet packets for image sequence coding

A new method for image sequence coding involving a wavelet packets transform is proposed to compact the energy of the transformed signal while preserving acceptable restitution quality. Based on an interframe coding scheme, the sequence is split into groups of frames where a similarity between successive frames exists. The best basis is generated for the first intra-frame, the first inter-predicted frame and the first inter-interpolated frame of each group and conserved for the following corresponding frames to reduce the computational complexity. Two different methods for on-line selection of the best basis are presented. The choice of the proposed technique is validated by experimental results where a codec based on motion compensated wavelet packets transform is presented.

Nonlinear optical flow estimation and segmentation

We present an algorithm for the joint estimation and segmentation of optical flow from gradient data, based on considerations from image grey level segmentation. Some first simulation results are shown and compared to existing spatial domain flow estimation algorithms. Promising results as well as strong indications for future enhancements are obtained.