Manfred Feil

Multicomputer algorithms for wavelet packet image decomposition

In this paper we describe and analyze algorithms for 2-D wavelet packet decomposition for MIMD distributed memory architectures. We discuss two different approaches: On the one hand algorithms generating the entire wavelet packet subband structure (as required for adaptive applications), on the other hand algorithms generating the lowest subband level only (as required for numerical applications). We investigate several optimizations and generalizations of corresponding message passing algorithms and finally compare the results obtained on a Cray T3D and a Parsytec GCel 1024.

Algorithms and Programming Paradigms for 2-D Wavelet Packet Decomposition on Multicomputers and Multiprocessors

In this work we describe and analyze algorithms for 2-D wavelet packet decomposition for multicomputers and multiprocessors. In the case of multicomputers we especially focus on the question of handling of the border data among the processing elements. For multiprocessors we discuss several optimizations of data parallel algorithms and finally we compare the results obtained on a multiprocessor employing the message passing and data parallelism paradigm, respectively.

Wavelet packet zerotree image coding on multicomputers

In this paper, we describe and analyze algorithms for advanced image coding on distributed-memory MIMD architectures. In particular, we consider a wavelet packet-based image codec using the concept of zerotree encoding. The main contribution of this work is the design of a parallel image coder composed of a wavelet packet decomposition in conjunction with the best basis algorithm followed by zerotree coding. Whereas two sensible parallelization techniques can be employed for the wavelet packet decomposition (subband-based partitioning and stripe partitioning), the zerotree coding step only allows one reasonable parallelization method (stripe partitioning). We investigate the advantages and drawbacks of all combinations and show experimental results obtained on a Siemens hpcLine cluster and a Cray T3E.

2-D wavelet packet decomposition on multicomputers

In this work we describe and analyze algorithms for 2-D wavelet packet decomposition for MIMD distributed memory architectures. The main goal is the generalization of former parallel WP algorithms which are constrained to a number of processor elements equal to a power of 4. We discuss several optimizations and generalizations of data parallel message passing algorithms and finally compare the results obtained on a Cray T3D.

Wavelet packet image decomposition on MIMD Architectures

In this work, we describe and analyze algorithms for 2D wavelet packet (WP) decomposition for multicomputers and multiprocessors. In the case of multicomputers, the main goal is the generalization of former parallel WP algorithms which are constrained to a number of processor elements equal to a power of 4. For multiprocessors, we discuss several optimizations of shared-memory algorithms and finally we compare the results obtained on multicomputers and multi-processors employing the message passing (MPI) and shared-memory programming (OpenMP) paradigm, respectively.

Real-Time Image Analysis Using MIMD Parallel à trous Wavelet Algorithms

The a trous algorithm represents a discrete approach to the classical continuous wavelet transform. Similar to the fast or pyramidal wavelet transform, the input signal is analysed by using the coefficients of a properly chosen low-pass filter, but in contrast to the latter, all frequency sub bands are retained with full resolution. Therefore, this algorithm is much more demanding in terms of computational complexity compared to the fast wavelet transform and requires some sort of acceleration in order to satisfy real-time constraints. In this paper we develop parallel algorithms for different MIMD architectures for the two-dimensional a trous decomposition. In particular, classical border treatment strategies are discussed and compared in the context of data partitioning. It turns out that in contrast to the fast wavelet transform, the proper choice of a border treatment strategy does not depend on the underlying hardware. Additionally, only low scalability is achieved on multi-computers and multi-processors when employing the message passing paradigm, whereas much better behaviour is observed on multi-processors using the shared memory programming model.

Wavelet image and video coding on parallel architectures

We discuss parallel algorithms for wavelet-based image and video coding. After reviewing fundamentals of the parallel discrete wavelet transform, we cover the parallelization of two state-of-the-art compression schemes: a C++ 3D SPIHT video codec and a Java JPEG-2000 implementation.

Computation of the Continuous Wavelet Transform on Massively Parallel SIMD Arrays

Strategies for computing the continuous wavelet transform on massively parallel SIMD arrays are introduced and discussed. The different approaches are theoretically assessed and the results of implementations on a MasPar MP-2 are compared.

Motion-compensated wavelet packet zerotree video coding on multicomputers

In this work we describe and analyze algorithms for advanced video coding on distributed memory MIMD architectures. In particular, we consider a wavelet packet based codec using the concept of zerotree encoding. The main contribution of this work is the design of a parallel motion-compensated video coder composed of a wavelet packet decomposition in conjunction with the best basis algorithm followed by zerotree coding. Whereas two sensible parallelization techniques can be employed for the wavelet packet decomposition (subband based partitioning and stripe partitioning), the zerotree coding and motion compensation stages only allow one reasonable parallelization method (stripe partitioning). We investigate the advantages and drawbacks of the resulting different overall data distribution strategies and show experimental results obtained on a Siemens hpcLine cluster and a Cray T3E.

Parallel Wavelet Transforms on Multiprocessors

We discuss several issues relevant for parallel wavelet transforms and their possible implications on the choice of a proper programming paradigm for corresponding multiprocessor implementations.