Daniel Hofman

Unified, multiple target, computer engineering learning platform

To overcome a complex and multidisciplinary approach which includes understanding of various systems based on different technologies and system solution optimizations in engineering education for embedded systems at university level and provide a complex and multidisciplinary approach, the first cycle of the design of a unified, multiple-target platform that will enable integrated approach to computer engineering and embedded systems learning has been finished so far. This work shows results of current research, preliminary conclusions and future focuses.

Novel Prediction Based Technique for Efficient Compression of Medical Imaging Data

The annual volume of imaging data in modern paperless hospitals can approach up to 10 terabytes, heavily pressing the storage and transmission requirements. Utilizing efficient compression techniques for those data in order to reduce associated costs is very attractive from both viewpoints: financial and organizational. Although lossy techniques can yield better compression results, due to possible compression artifacts in the compressed image, they are less favored compared to lossless compression techniques in certain medical applications such as image-based diagnosis, archival etc. Moreover, new approaches in medical imaging such as 3D and 4D imaging and bio–modeling produce even greater amounts of image data. For efficient storage and transmission of those data and utilization of systems that exploit 3D and 4D imaging technologies, compression is inevitable. In this field, at least certain parts of images are required to be stored and transmitted without any loss of information. The lossless compression algorithm that we propose can also be efficiently employed for at least those vital parts of interest in this kind of applications. We propose a higly adaptive prediction-based lossless compression algorithm which models nontrivial image structures through selective blends of static predictors.

3D Object Classification and Segmentation Methods

Future multimedia high-quality systems will be, among all, based on improving 3D visual experience. To raise 3D visual content quality and interactivity it is necessary to enable segmentation and classification of content which involves dividing the scene into meaningful sub-regions with the same attributes. Partitioning the image into grouping objects has various different applications in a wide variety of areas, since distinctive features in raw images may appear unclear to the human eyes. Segmentation can be defined as the identification of meaningful image components. It is a fundamental task in image processing providing the basis for any kind of further high-level image analysis. There are many different ways of segmenting the 3D image, all of which can be considered as a good segmentations, depending on objects of interest on an image, and to a large extent, the user’s own subjectivity. Key issues in this chapter include different techniques for segmentation of 3D object based on classification on different regions and shapes.

Power consumption and bandwidth savings with video transcoding to mobile device-specific spatial resolution

Usage of mobile devices for multimedia content playback is drowning the battery power rapidly. High quality videos which are streamed from the internet are power consuming and demand high network bandwidth. We measured the power consumption and bitrate of video sequences in respect to different spatial resolutions and used acceptability-based Quality of Experience (QoE) model for determining the impact of video resolution on QoE. The results showed that both power and bandwidth can be saved without noticeable reduction in QoE when transcoding video to a resolution specific to the mobile device instead of using standard video resolutions.

Exploring manycore architectures for next-generation HPC systems through the MANGO approach

Abstract The Horizon 2020 MANGO project aims at exploring deeply heterogeneous accelerators for use in High-Performance Computing systems running multiple applications with different Quality of Service (QoS) levels. The main goal of the project is to exploit customization to adapt computing resources to reach the desired QoS. For this purpose, it explores different but interrelated mechanisms across the architecture and system software. In particular, in this paper we focus on the runtime resource management, the thermal management, and support provided for parallel programming, as well as introducing three applications on which the project foreground will be validated.

Work in progress: Embedded computer engineering learning platform capabilities

Provided unified learning platform, which is developed as a main goal of our ICT FP7 - Collaborative Project: Embedded Computer Engineering Learning Platform enables modular approach in education of computer engineers. It helps engineering education personnel to transform passive listeners students into active learners, thus stimulating students to actively participate in the learning process. Furthermore, this platform shall introduce a flexible and extendable learning environment for upcoming technologies in embedded systems, thus providing a long lasting educational solution for academia. This work in progress article describes capabilities of our learning platform.

Parallelizing MPEG Decoder with Scalable Streaming Computation Kernels

In this paper, we describe a scalable and portable parallelized implementation of a MPEG decoder using a streaming computation paradigm, tailored to new generations of multi—core systems. A novel, hybrid approach towards parallelization of both new and legacy applications is described, where only data—intensive and performance—critical parts are implemented in the streaming domain. An architecture—independent StreamIt language is used for design, optimization and implementation of parallelized segments, while the developed StreamGate interface provides a communication mechanism between the implementation domains. The proposed hybrid approach was employed in re—factoring of a reference MPEG video decoder implementation; identifying the most performance—critical segments and re—implementing them in StreamIt language, with StreamGate interface as a communication mechanism between the host and streaming kernel. We evaluated the scalability of the decoder with respect to the number of cores, video frame formats, sizes and decomposition. Decoder performance was examined in the presence of different processor load configurations and with respect to the number of simultaneously processed frames.