Joerg Heuer

Global motion estimation in image sequences using robust motion vector field segmentation

In this paper we propose an algorithm for the purpose of video indexing which estimates the global motion caused by camera movement. Since motion estimation is a computationally expensive process, we are using information already computed in block based video encoders such as the displacement vectors of the blocks and the computed SAD values of the blockmatching process. With this approach we could significantly speed up the determination of the parameters of a chosen global motion model. The accuracy of the algorithm is improved further by introducing a reliability measure. To achieve a robust computation with respect to object motion, an effective and straightforward segmentation of this vector field is implemented. Finally, the modeling of the global motion is evaluated using MPEG-7 test data.

Coding format independent multimedia content adaptation using XML

Due to the heterogeneity of the current terminal and network infrastructures, multimedia content needs to be adapted to specific capabilities of these terminals and network devices. Furthermore, user preferences and user environment characteristics must also be taken into consideration. The problem becomes even more complex by the diversity of multimedia content types and encoding formats. In order to meet this heterogeneity and to be applicable to different coding formats, the adaptation must be performed in a generic and interoperable way. As a response to this problem and in the context of MPEG-21, we present an approach which uses XML to describe the high-level structure of a multimedia resource in a generic way, i.e., how the multimedia content is organized, for instance in layers, frames, or scenes. For this purpose, a schema for XML-based bitstream syntax descriptions (generic Bitstream Syntax Descriptions or gBSDs) has been developed. A gBSD can describe the high-level structure of a multimedia resource in a coding format independent way. Adaptation of the resource is based on elementary transformation instructions formulated with respect to the gBSDs. These instructions have been separated from the gBSDs in order to use the same descriptions for different adaptations, e.g., temporal scaling, SNR scaling, or semantic adaptations. In the MPEG-21 framework, those adaptations can be steered for instance by the network characteristics and the user preferences. As a result, it becomes possible for coding format agnostic adaptation engines to transform media bitstreams and associated descriptions to meet the requirements imposed by the network conditions, device capabilities, and user preferences.

BeTrIS: an index system for MPEG-7 streams

The ISO/IEC Motion Picture Group (MPEG) issued in 2002 a standard, called MPEG-7, which enables the content description of multimedia data in XML. The standard supports applications to exchange, identify, and filter multimedia contents based on MPEG-7 descriptions. However, especially mobile applications that deal with MPEG-7 suffer from limited bandwidth, low computational power, and limited battery life. In this document, we describe an index system adopted from database systems that allows filter mechanisms and random access to encoded MPEG-7 streams and which overcome the limitation of the network and the consuming terminal. Encoding is applied in order to reduce the data rate of the XML documents to be transmitted. The indexed parts of the encoded streams can be accessed without the need to deserialize the complete stream. Furthermore, the system is evaluated and results of the experimental evaluation are discussed.

Visual feature discrimination versus compression ratio for polygonal shape descriptors

In the last decade several methods for low level indexing of visual features appeared. Most often these were evaluated with respect to their discrimination power using measures like precision and recall. Accordingly, the targeted application was indexing of visual data within databases. During the standardization process of MPEG-7 the view on indexing of visual data changed, taking also communication aspects into account where coding efficiency is important. Even if the descriptors used for indexing are small compared to the size of images, it is recognized that there can be several descriptors linked to an image, characterizing different features and regions. Beside the importance of a small memory footprint for the transmission of the descriptor and the memory footprint in a database, eventually the search and filtering can be sped up by reducing the dimensionality of the descriptor if the metric of the matching can be adjusted. Based on a polygon shape descriptor presented for MPEG-7 this paper compares the discrimination power versus memory consumption of the descriptor. Different methods based on quantization are presented and their effect on the retrieval performance are measured. Finally an optimized computation of the descriptor is presented.