Roland Tusch

VIDEX : an integrated generic video indexing approach

This paper presents an integrated generic technique for low-and high-level video indexing. The proposed approach tries to integrate the advantages of existing low- and high-level video indexing approaches by reducing their shortcomings. Furthermore, the model introduces concepts for a detailed structuring of video streams, and for correlations of low-and high-level video objects. The proposed model is called generic, as it only defines a framework of classes for an implementating a prototype of a distibuted multimedia information system supporting content-based video retrieval.

Context-Aware UPnP-AV Services for Adaptive Home Multimedia Systems

One possibility to provide mobile multimedia in domestic multimedia systems is the use of Universal Plug and Play Audio Visual (UPnP-AV) devices. In a standard UPnP-AV scenario, multimedia content provided by a Media Server device is streamed to Media Renderer devices by the initiation of a Control Point. However, there is no provisioning of context-aware multimedia content customization. This paper presents an enhancement of standard UPnP-AV services for home multimedia environments regarding context awareness. It comes up with context profile definitions, shows how this context information can be queried from the Media Renderers, and illustrates how a Control Point can use this information to tailor a media stream from the Media Server to one or more Media Renderers. Moreover, since a standard Control Point implementation only queries one Media Server at a time, there is no global view on the content of all Media Servers in the UPnP-AV network. This paper also presents an approach of multimedia content integration on the Media Server side that provides fast search for content on the network. Finally, a number of performance measurements show the overhead costs of our enhancements to UPnP-AV in order to achieve the benefits.

A Mobile Agent-Based Infrastructure for an Adaptive Multimedia Server

This paper introduces a mobile agent-based infrastructure for an adaptive multi-media server enabling a dynamic migration or replication of certain multimedia applications among a set of available server nodes. It discusses the requirements from both, the server’s and the middleware’s point of view to each other and comes up with a specification and implementation of a CORBA-based interface between them.

Robust Traffic State Estimation on Smart Cameras

This paper presents a novel method for video-based traffic state detection on motorways performed on smart cameras. Camera calibration parameters are obtained from the known length of lane markings. Mean traffic speed is estimated from Kanade-Lucas-Tomasi (KLT) optical flow method using a robust outlier detection. Traffic density is estimated using a robust statistical counting method. Our method has been implemented on an embedded smart camera and evaluated under different road and illumination conditions. It achieves a detection rate of more than 95% for stationary traffic.

Feature-based level of service classification for traffic surveillance

A novel level of service (LOS) estimation approach based on the extraction of three local visual features is presented. The feature set comprises KLT motion vectors and Sobel edges, and is fed into a Gaussian radial-basis-function (GRBF) network to classify the prevailing LOS. The whole approach is designed and implemented to run on smart cameras in real-time and has been evaluated with a comprehensive set of real-world training and test video data from a national motorway. The evaluations in daylight environments have shown an average accuracy of LOS classification of 86.2% on an Atom-based smart camera, with a maximum reachable processing frame rate of 12.5 frames/sec. Incorrect classified samples differed from the ground truth by only one level. The comparisons are done with observation data from sensors utilizing a combination of Doppler radar, ultrasound, and passive infrared technologies.

Metadata Integration and Media Transcoding in Universal-Plug-and-Play (UPnP) Enabled Networks

Universal plug and play (UPnP) is a widely accepted standard for automatically detecting devices and services in a local area network as well as for describing and controlling them. In order to deal with multimedia devices and especially content, in 2002 the UPnP-AV standard definition was released. It defines device and service descriptions for media servers and renderers. Thereby, the media servers content directory service allows an easy management and the exchange of metadata about the provided media data. Media content became browsable by semantic meta information about it. There are still two major drawbacks of UPnP-AV, which make its usage in real world multimedia communication scenarios very difficult. First, searching for similar content on distributed media servers with a huge number of media files is not economically possible. Second, the media content must be consumed by Renderers as provided by the Servers, independently of their terminal capabilities and network connections. In order to deal with these two drawbacks, this work proposes a novel approach of metadata integration and media transcoding in UPnP networks. First, the Media Server is extended by a control point which offers discovery of other media servers and fetches metadata from their content directories. Furthermore, it integrates the gathered information in its own content directory. Control Points are then able to query this Integrating Media Server for a desired content, and get a network-complete search result. Second, terminal and network capabilities of the Renderers are taken into account in order to transcode and transmit the content in a suitable way for the consuming device. These two approaches of metadata integration and media data adaptation enable searchable logical views on tailored multimedia content in UPnP-AV networks.

The SMOOTH video DB - demonstration of an integrated generic indexing approach

The SMOOTH Video DB is a distributed system proposing an integral query, browsing, and annotation software framework in common with an index database for video media material.

Efficient Level of Service Classification for Traffic Monitoring in the Compressed Video Domain

This paper presents a new method for estimating the level of service (LOS) on motorways in the compressed video domain. The method performs statistical computations on motion vectors of MPEG4 encoded video streams within a predefined region of interest to determine a set of four motion features describing the speed and density of the traffic stream. These features are fed into a Gaussian radial basis function network to classify the corresponding LOS. To improve the classification results, vectors of moving objects are clustered and outliers are eliminated. The proposed method is designed to be executed on a server system, where a large number of camera live streams can be analyzed in parallel in real-time. Evaluations with a comprehensive set of real-world training and test data from an Austrian motorway have shown an average accuracy of 86.7% on the test data set for classifying all four LOS levels. With a mean execution time of 48 microseconds per frame on a common server, hundreds of video streams can be analyzed in real-time.

LOOK2 - A Video-based System for Real-time Notification of Relevant Traffic Events

We demonstrate our novel video-based real-time traffic event notification and verification system LOOK2. It generates fast and reliable traffic information about relevant traffic state and road conditions changes on observed roads. It utilizes installed road-side sensors providing low-level traffic and environmental data, as well as video sensors which gain high-level traffic information from live video analysis. Spatio-temporal data fusion is applied on all available traffic and environmental data to gain reliable traffic information. This traffic information is published by a DATEXII compliant web service to a web-based traffic desk application. Road network and traffic channel operators receive real-time and relevant traffic event notifications by using this application. The system also enables a visual verification of the notified situations.

AVSS 2011 demo session: Level of service classification for smart cameras

Summary form only given. Automated code analysis is technology aimed at locating, describing and repairing areas of weakness in code. Code weaknesses range from security vulnerabilities, logic errors, concurrency violations, to improper resource usage, violations of architectures or coding guidelines. Common to all code analysis techniques is that they build abstractions of code and then check those abstractions for properties of interest. For instance a type checker computes how types are used, abstract interpreters and symbolic evaluators check how values flow, model checkers analyze how state evolves. Building modern program analysis tools thus requires a multi-pronged approach to find a variety of weaknesses. In this talk I will discuss and compare several program analysis tools, which MSR build during the last ten years. They include theorem provers, program verifiers, bug finders, malware scanners, and test case generators. I will describe the need for their development, their innovation, and application. Many of these tools had considerable impact on Microsofts development practices, as well as on the research community. Some of them are being shipped in products such as the Static Driver Verifier or as part of Visual Studio. Performing program analysis as part of quality assurance is meanwhile standard practice in many software development companies. However several challenges have not yet been resolved. Thus, I will conclude with a set of open challenges in program analysis which hopefully triggers new aspiring directions in our joint quest of delivering predictable software that is free from defect and vulnerabilities.