Claudia Schremmer

Empirical evaluation of layered video coding schemes

Video coding for Internet applications faces major challenges. Due to the heterogeneity of the network, users with very different access bandwidths to the Internet want to receive videos of the best possible quality. A good coding scheme for layered video pursues the following major goal: it maximises the subjective visual quality at a given bandwidth. In this article, we present four different techniques for video layering. The work focuses on an evaluation carried out on 30 test probands. The objective was twofold: finding both a suitable metric for automatic video quality assessment, as well as the best layering technique. The experimental results lead us to pronounce recommendations on the metric for automatic video quality assessment, and on a best video layering technique with regard to human perception.

Video-scaling algorithm based on human perception for spatiotemporal stimuli

Video conferencing and high quality video-on-demand services are very desirable for many Internet users. However, Internet access channels ranging from wireless connections to high- speed ATM networks mean great heterogeneity with respect to bandwidth. Hierarchical video encoders that scale and distribute video data over different layers enable users to adapt video quality to the capacity of their Internet connection. However, the construction of the layers at the encoder determines the video quality that can be expected at the receiver. To achieve an optimal configuration of the different layers with respect to visual quality, we propose a hybrid scaling algorithm that scales video data both in spatial and temporal dimension. Using a quality metric based on properties of the human visual system our algorithm calculates an optimal ratio between spatial and temporal information. Additionally, we present experimental results that demonstrate the capabilities of our approach.

Decomposition strategies for wavelet-based image coding

The wavelet transform has become the most interesting new algorithm for still image compression. Yet there are many parameters within a wavelet analysis and synthesis which govern the quality of a decoded image. In this paper, we discuss different decomposition strategies of a two-dimensional signal and their implications for the decoded image: a pool of gray-scale images has been wave let-transformed with different settings of the wavelet filter bank, quantization threshold and decomposition method. Contrary to the new standard JPEG-2000, where nonstandard decomposition is implemented, our investigation proposes standard decomposition for low-bitrate coding.

Empirical Evaluation of Boundary Policies for Wavelet-Based Image Coding

The wavelet transform has become the most interesting new algorithm for still image compression. Yet there are many parameters within a wavelet analysis and synthesis which govern the quality of a decoded image. In this paper, we discuss different image boundary policies and their implications for the decoded image. A pool of gray-scale images has been wavelet-transformed at different settings of the wavelet filter bank and quantization threshold and with three possible boundary policies.Our empirical evaluation is based on three benchmarks: a first judgment regards the perceived quality of the decoded image. The compression rate is a second crucial factor. Finally, the best parameter settings with regard to these two factors is weighted with the cost of implementation. Contrary to the JPEG2000 standard, where mirror paddingis implemented, our investigation proposes circular convolution as the boundary treatment.

Erfahrungen mit synchronen und asynchronen Lernszenarien an der Universität Mannheim

ZUSAMMENFASSUNG An der Universität Mannheim werden seit 1996 TeleTeaching-Projekte durchgeführt. Der Schwerpunkt liegt dabei in synchronen Lernszenarien, worunter wir eine zeitgleiche (Internet-) Übertragung einer Veranstaltung an verschiedene Orte verstehen. Die Audio- und Videoströme des Dozenten werden zugleich aufgezeichnet und als Bestandteil eines Computer-Based-Trainings zeitunabhängig für asynchrone Lernszenarien zur Verfügung gestellt. Die implementierten synchronen und asynchronen Lernszenarien werden seit 1998 im Verbundprojekt VIROR der oberrheinischen Universitäten Freiburg, Heidelberg, Karlsruhe und Mannheim eingesetzt. Dieser Artikel beschreibt verschiedene Lernszenarien bei TeleVeranstaltungen, ihre Besonderheiten und spezifischen Anforderungen. Die zugrunde liegende Technologie wird kurz vorgestellt, einige selbst entwickelte Software Werkzeuge beschrieben und weitere, bislang nicht realisierte Anforderungen an die Technik formuliert. Das Hauptaugenmerk aber legen wir auf Erfahrungen, die wir in den nunmehr vier Jahren aktiven TeleTeachings gewonnen haben.

Erzeugung und Verwendung multimedialer Teachware im synchronen und asynchronen Teleteaching

Authoring multimedia teachware is a costly process. In traditional teaching environments a number of documents are created to meet the requirements of different teaching scenarios (e. g. lectures, student homework) but deal with the same content. Furthermore, explanations of a teacher during a lecture implicitly form a document that may be useful beyond the boundaries of the lecture. In this paper we describe how documents created during a lecture can be recorded and how these recordings can be made available on a server in the Internet. The presented scheme allows them to be linked to additional multimedia documents, making a Computer Based Training unit. Thus, our scheme enables the efficient use of teachware in synchronous as well as in asynchronous teaching scenarios.