Jack Kouloheris

Improving server utilization using fast virtual machine migration

Live virtual machine (VM) migration is a technique for transferring an active VM from one physical host to another without disrupting the VM. In principle, live VM migration enables dynamic resource requirements to be matched with available physical resources, leading to better performance and reduced energy consumption. However, in practice, the resource consumption and latency of live VM migration reduce these benefits to much less than their potential. We demonstrate how these overheads can be substantially reduced, enabling live VM migration to fulfill its promise. Specifically, we first experimentally study several factors that contribute to the resource consumption and latency of live VM migration, including workload characteristics, the hypervisor and migration configuration, and the available system and network resources. Then, from the insights gained, we propose an alternative remote direct memory access-based migration technique that significantly reduces VM migration overheads. Finally, via simulation and experiments with real system prototypes, we demonstrate that the reduced VM migration overhead results in significant improvements in resource and energy efficiencies, relative to existing migration techniques.

Full-reference video quality assessment considering structural distortion and no-reference quality evaluation of MPEG video

There has been an increasing need recently to develop objective quality measurement techniques that can predict perceived video quality automatically. This paper introduces two video quality assessment models. The first one requires the original video as a reference and is a structural distortion measurement based approach, which is different from traditional error sensitivity based methods. Experiments on the video quality experts group (VQEG) test data set show that the new quality measure has higher correlation with subjective quality evaluation than the proposed methods in VQEGs Phase I tests for full-reference video quality assessment. The second model is designed for quality estimation of compressed MPEG video stream without referring to the original video sequence. Preliminary experimental results show that it correlates well with our full-reference quality assessment model.

Foveated wavelet image quality index

The human visual system (HVS) is highly non-uniform in sampling, coding, processing and understanding. The spatial resolution of the HVS is highest around the point of fixation (foveation point) and decreases rapidly with increasing eccentricity. Currently, most image quality measurement methods are designed for uniform resolution images. These methods do not correlate well with the perceived foveated image quality. Wavelet analysis delivers a convenient way to simultaneously examine localized spatial as well as frequency information. We developed a new image quality metric called foveated wavelet image quality index (FWQI) in the wavelet transform domain. FWQI considers multiple factors of the HVS, including the spatial variance of the contrast sensitivity function, the spatial variance of the local visual cut-off frequency, the variance of human visual sensitivity in different wavelet subbands, and the influence of the viewing distance on the display resolution and the HVS features. FWQI can be employed for foveated region of interest (ROI) image coding and quality enhancement. We show its effectiveness by using it as a guide for optimal bit assignment of an embedded foveated image coding system. The coding system demonstrates very good coding performance and scalability in terms of foveated objective as well as subjective quality measurement.

High-performance video server for broadband network environment

We present a multimedia server comprising of multiple stream controllers connected to a host (work station) I/O bus. Each stream controller manages an array of disks in which multimedia data is stored in format of network packets. When the host instructs the stream controller to deliver a multimedia file to a client, the stream controller retrieves the network packets of that file, completes the missing header/trailer fields, and forwards these packets to the network interface directly, avoiding any further involvement of the host. To avoid interference on the disks, data is interleaved across all disks connected to a stream controller in fixed play back time units. This helps reduce the jitter in the response time of the disks, and therefore, the size of the buffers needed to maintain interruption free delivery. Metadata is stored with the network packets of a video/multimedia file to enable the stream controller to autonomously fetch a complete multimedia/video file without host intervention. In a departure from traditional RAID, the stream controller simultaneously issues a set of read commands periodically, one for each active multimedia stream. Each reach command retrieves a full interleave unit from a single disk, and the set of simultaneously issued read commands are distributed across all disks.

Statistical model based video segmentation and its application to very low bit rate video coding

We present a statistical model-based video segmentation algorithm for typical videophone and video-conference applications. This algorithm makes use of online information and tries to build statistical models for both background and foreground and update the models on the fly. A hierarchical system structure is designed and segmentation is combined with tracking. Two possible applications are discussed: to generate VOP for MPEG-4 and to introduce subjective rate control for DCT-based algorithms. A rate distortion based rate control algorithm for H.263 is proposed and implemented as an example.

Adaptive frame prediction for foveation scalable video coding

Embedded rate scalable video coding allows for the extraction of coded visual information at continuously varying bit rates from a single compressed bitstream. This is a very attractive feature for many multimedia communication applications. Motion estimation (ME) /motion compensation (MC) techniques are widely employed in various video coding systems to reduce temporal information redundancy. One of the major challenging problems in ME/MC based rate scalable video coding is how to generate the prediction frame from the previous frame to match the current frame. This problem is more difficult in rate scalable coding than in fixed rate coding because the decoding data rate is unavailable to the encoder. We propose an adaptive frame prediction scheme for foveation scalable video coding (FSVC), which is a new video coding algorithm that combines a foveation-based human visual system (HVS) model with a wavelet-based rate scalable coding algorithm. The new frame prediction algorithm provides an adaptive mechanism to control the prediction errors while reduce error propagation.

Efficient and low-cost video transcoding

We present a low cost high performance scheme for video transcoding. Our scheme includes the information look-ahead structure and the transcoding algorithm. It is this look-ahead mechanism that greatly facilitates the use of information embedded in the bitstream for informative and intelligent transcoding. Furthermore, our transcoding algorithm has four new features: 1) picture complexity look-ahead for efficient rate allocation and control; 2) dynamic bit allocation weighting on picture types based on the scene complexity and bit rate; 3) adaptive nonlinear rate control on macroblocks based on complexity distribution in the picture; 4) Intra frame requantization adjustment to reduce the transcoding distortion. The software simulation results have proven that the performance of our transcoding scheme is significantly better than the reencoding scheme and comparable to the direct encoding scheme.

Low-cost and efficient logo insertion scheme in MPEG video transcoding

In this paper, we work on the problem of inserting logo into part of a previously compressed video sequence while transcoding it to a video stream with lower rate. In most of the transcoding architecture, it has been widely accepted to reuse the motion information extracted from the original incoming bit stream. For our transcoding and logo insertion process, we preserve the same motion information as the incoming video stream for the parts that were unaffected by logo, and propose an efficient algorithm to code logo areas. We first find the proper range of pictures that might be affected by logo insertions, analyze logo-affected parts of those pictures, and perform corresponding algorithms depending on the picture types. In addition, we develop a method to achieve better bit allocation during logo insertion. The efficiency of our algorithm is demonstrated by simulation.

A cost function with position penalty for motion estimation in MPEG-2 video coding

We propose a new cost function for the motion estimation algorithm. A position penalty has been added to the conventional cost function, Sum of Absolute Difference (SAD), to regulate the motion field. The effort is focused on minimization of the size of motion vector difference as well as residual error which need to be coded. Compared to the existing motion field regulation methods, the proposed method has a better chance to choose minimum distortion which allows it to have better picture quality with a fixed bit rate. Simulation results show that the proposed cost function outperforms the existing methods with minimal increase of computational cost, which makes it attractive for hardware implementation.

Statistical model-based video segmentation and its application to very low bit-rate video coding

This paper presents a statistical model-based video segmentation algorithm for typical videophone and videoconference applications. This algorithm makes use of online information to build and track statistical models for both the background and foreground on the fly. The segmentation algorithm is then rendered as a MAP problem. A hierarchical system structure is designed and spatial and temporal filters are used to improve the segmentation quality. The algorithm is implemented on a PC and runs in real time. In addition, two possible applications are discussed: generating video objects for the upcoming MPEG-4 standard and introducing subjective factors into the rate control of DCT-based coding algorithms. We focus on the second application by proposing a rate-distortion (R-D)-based optimal rate control algorithm for H.263. In this rate control algorithm, which we refer to as region-based rate control algorithm, previous segmentation results are used as subjective knowledge. A distortion model is created to integrate both subjective and objective factors and an R-D criterion is used to obtain the optimal bit allocation. With the proposed algorithm, an H.263 compatible encoder is implemented and it produces better perceptual quality in our experiments than standard H.263.