This paper presents a framework for automatic video region-of-interest determination based on visual attention model. We view this work as a preliminary step towards the solution of high-level semantic video analysis. Facing such a challenging issue, in this work, a set of attempts on using video attention features and knowledge of computational media aesthetics are made. The three types of visual attention features we used are intensity, color, and motion. Referring to aesthetic principles, these features are combined according to camera motion types on the basis of a new proposed video analysis unit, frame-segment. We conduct subjective experiments on several kinds of video data and demonstrate the effectiveness of the proposed framework.

In this letter, we study a hybrid Multimedia-on-Demand (MoD) system which provides broadcast, batch and interactive services. An analytical model for such an MoD system is provided. Numerical results show that with proper design, the system can provide better performance than those systems which only provide any subset of two services.

Future wireless/mobile system is expected to have heterogeneous wireless overlay networks for ubiquitous multimedia communication. In a such network environment, mobile users are likely to try to get attached to higher bandwidth network as bandwidth-hungry multimedia applications are increasing. However, the users have to perform vertical handoff to lower bandwidth network, as high bandwidth network become unavailable due to various reasons (such as its limited coverage, network congestion, etc.). In this paper, we discuss the problem of vertical handoff from a user's perspective. For this purpose, we formulate user satisfaction as a function of bandwidth utility and handoff latency. Then, we investigate the effect of call holding time, user movement probability, etc. on the satisfaction that a user derives from the use of network service for multimedia applications. In addition, based on the evaluation, we present an algorithm for selecting a wireless network, which maximizes the effective user satisfaction.

The technological development of wireless environment has made real-time multimedia communications possible in wireless networks. Many studies have been done on real-time communications in wireless networks in order to overcome a higher bit error rate in wireless channels. However, none of work deals with firm real-time communications which can be applied to multimedia communications. In this paper, we propose an adaptive error correcting scheme for firm real-time multimedia communications in wireless networks in order to maximize the expected net profit. The proposed scheme adaptively selects an error correcting code under the current air state and the message state of a message stream. Throughout simulation results, we show that the suggested scheme provides more profit than single error-correcting code schemes.

With reference to video motion estimation in the framework of the new H.264/AVC video coding standard, this paper presents algorithmic and architectural solutions for the implementation of context-aware coprocessors in real-time, low-power embedded systems. A low-complexity context-aware controller is added to a conventional Full Search (FS) motion estimation engine. While the FS coprocessor is working, the context-aware controller extracts from the intermediate processing results information related to the input signal statistics in order to automatically configure the coprocessor itself in terms of search area size and number of reference frames; thus unnecessary computations and memory accesses can be avoided. The achieved complexity saving factor ranges from 2.2 to 25 depending on the input signal while keeping unaltered performance in terms of motion estimation accuracy. The increased efficiency is exploited both for (i) processing time reduction in case of software implementation on a programmable platform; (ii) power consumption reduction in case of dedicated hardware implementation in CMOS technology.

Radio resource is the bottleneck for current multimedia wireless networks. Intelligent traffic control strategies can be enforced to optimize resource allocation so as to enhance network performance. In this study, dynamic control scheme for non-real-time traffic and autonomic control schemes for multimedia traffic are proposed to guarantee the required quality of service (QoS) in the inference-dominated high-speed wireless environment. Both handoff priority and terminal mobility are also taken into consideration. The performance of the state-dependent multidimensional birth-death process is derived by the efficient matrix-analytic methods (MAMs). Compared with the previous results, this paper shows that the proposed control methods can be used for both real-time and non-real-time multimedia traffic in order to meet the required performance without degrading the quality of multimedia services. These results are also important for the design of evolving multimedia wireless systems as well as network optimization.

To overcome the mobility impact on RSVP, many schemes have been proposed based on Mobile IP regional registration and passive reservation in advance. Although the regional registration and in advance reservation reduce the QoS interruption time, the latter may demand intolerable bandwidth. This letter introduces a novel approach to reduce the QoS interruption time by maximizing the localization of QoS re-establishment in the regional registration environment. The proposed scheme identifies the exact path segment affected by mobility. The QoS interruption time of the proposed scheme is comparable to its low bound without in advance reservation.

One of the important problems in overlay multicast is how to deal with node failures and ungraceful leavings. When a non-leaf end host fails or leaves the multicast session, all downstream nodes will be affected. In this paper, we adopt the proactive approach, which pre-calculates a candidate node (called parent-to-be) for each node to connect to in case its current parent dies. The goal is to recover the overlay multicast tree quickly so that the disruption of service to those affected nodes is minimized. We combine the local parent-to-be locating and global parent-to-be locating schemes together, in order to take advantage of less interference in the local scheme and the flexibility of the global scheme. The quality of the recovered tree is improved while the responsiveness of the proactive approach is maintained.

This paper introduces the water ring scan method especially designed for the scalable video coding schemes such as fine granularity scalabilities (FGS) on the basis of MPEG-4 part-2 and the H.264. The proposed scanning method can improve the subjective quality of the decoded video by most-preferentially encoding, transmitting and decoding the image information of the region of interest. From the various simulation results of FGS coding schemes with MPEG-4 part 2 and H.264, the proposed scanning method can improve the subjective picture quality about 0.5 dB 3.5 dB better than the widely used raster scan order, especially on the region of interest, without significant loss of the quality in the left-over region.

In this study, we propose an object-based multimedia model for specifying the QoS (quality of service) requirements, such as the maximum data-dropping rate or the maximum data-delay rate. We also present a resource allocation model, called the net-profit model, in which the satisfaction of user's QoS requirements is measured by the benefit earned by the system. Based on the net-profit model, the system is rewarded if it can allocate enough resources to a multimedia delivery request and fulfill the QoS requirements specified by the user. At the same time, the system is penalized if it cannot allocate enough resources to a multimedia delivery request. We first investigate the problem of how to allocate resources efficiently, so that the QoS satisfaction is maximized. However, the net-profit may be distributed unevenly among the multimedia delivery requests. Thus, the second problem discusses how to allocate the resource efficiently so that the net-profit difference is minimized between any two multimedia requests. A dynamic programming based algorithm is proposed to find such an optimal solution with the minimum net-profit differences.

This paper introduces a new frame-based bit-rate control scheme for bandwidth-adaptive video coding. Proposed method can accurately adapt to the rapid varying scene characteristics by reducing the number of occurrences of the extrapolations while updating the rate-distortion model used for determine the appropriate quantization steps.

FEC (Forward Error Correction) is widely used to recover packet loss over the Internet since it does not involve additional network delay. However, FEC still needs much additional network bandwidth for redundancy, and does not consider the priority or the importance of video frames to generate redundant data. In this paper, we present Periodic FEC (PFEC) to make up for the shortcomings of FEC. PFEC divides frames into high-priority frames and low-priority frames, and gives redundancy only to high-priority frames. As specific examples, we describe two types of PFEC: Media-Independent PFEC and Media-Dependant PFEC. Moreover, based on the two-state continuous time Markov chain, we propose redundancy control algorithms of the PFEC schemes that can adjust the amount of redundancy to optimal levels depending on network loss conditions. For better performance, we also consider UEP (Unequal Error Protection) based on PFEC that gives redundancy to low-priority frames as well as high-priority frames. Experimental results show that compared with FEC, PFEC reduces the amount of redundancy considerably but degrades PSNR slightly, and UEP based on PFEC economizes redundancy without the degradation of the PSNR.

Next generation wireless networks are expected to support multimedia applications (audio phone, video on demand, video conference, file transfer, etc.). Multimedia applications make a great demand for bandwidth and impose stringent quality of service (QoS) requirements on the wireless networks. In order to provide mobile hosts with high QoS, efficient and better bandwidth reservation is necessary in multimedia wireless networks. This paper presents a novel hybrid dynamic-grouping bandwidth reservation scheme to support QoS guarantees in the next generation wireless networks. The proposed scheme is based on probabilistic resource estimation to provide QoS guarantees for multimedia traffic in cellular networks. We establish several reservation time-sections, called groups, according to the mobility information of mobile hosts (MHs) of each base station (BS). The amount of reserved bandwidth for each BS is dynamically adjusted for each reservation group. We use the hybrid dynamic-grouping bandwidth reservation scheme to decrease the connection-dropping probability (CDP) and connection-blocking probability (CBP), while increasing the bandwidth utilization. The simulation results show that the hybrid dynamic-grouping bandwidth reservation scheme provides less CDP and less CBP, and achieves high bandwidth utilization.

To fulfill the computing demands in visual media processing, we have been investigating a parallel processing system to improve the processing speed of the visual media related to applications from the point of view of a memory system within a single instruction multiple data (SIMD) computer. In this paper, we have introduced MAMS-PP4, which is similar to a pipelined SIMD architecture type and consists of pq processing elements (PEs) as well as a multi-access memory system (MAMS). MAMS supports simultaneous access to pq data elements within a horizontal (1 pq), a vertical (pq 1) or a block (p q) subarray with a constant interval in an arbitrary position in an M N array of data elements, where the number of memory modules, m, is a prime number greater than pq. MAMS reduces the memory access time for an SIMD computer and also improves the cost and complexity that involved in controlling the large volume of data demanded in visual media applications. PE is designed to be a two-state machine in order to utilize MAMS efficiently. MAMS-PP4 was fabricated into ASIC using TOSHIBA TC240C series library and a test board was used to measure the performance of ASIC. The test board consists of devices such as an MPC860 embedded-PCI board, two ASICs and a FPGA for the control units. Experiment was done on various computer systems in order to compare the performance of MAMS-PP4 using morphological operations as the application. MAMS-PP4 shows a respectful and consistent processing speed.

LOTOS parallel operator, which is a binary operator, is used to combine processes in order to express their concurrency. Unlike other LOTOS operators, various possibilities exist when combining processes by the parallel operator. If two processes are selected randomly for combining, the size of the composite intermediate process after combining may be large. In this paper, we propose an algorithm for selecting two processes out of three or more processes so that the size of the intermediate process is the smallest when combined by the parallel operator. Smaller size of an intermediate process means it takes less memory space which is very important in designing verification tools for systems or communication protocols specified in LOTOS.

This paper presents an application-level QoS comparison of three inter-destination synchronization schemes: the master-slave destination scheme, the synchronization maestro scheme, and the distributed control scheme. The inter-destination synchronization adjusts the output timing among destinations in a multicast group for live audio and video streaming over the Internet/intranets. We compare the application-level QoS of these schemes by simulation with the Tiers model, which is a sophisticated network topology model and reflects hierarchical structure of the Internet. The comparison clarifies their features and finds the best scheme in the environment. The simulation result shows that the distributed control scheme provides the highest quality of inter-destination synchronization among the three schemes in heavily loaded networks, while in lightly loaded networks the other schemes can have almost the same quality as that of the distributed control scheme.

In this paper, we propose the Synchronized Mobile Multicast (SMM) scheme for the real-time multimedia service to achieve three most important characteristics that the traditional Home Subscription (HS) and Remote Subscription (RS) mobile schemes cannot support. First, the SMM scheme supports the scalable one-to-many and many-to-many synchronized multimedia multicast on mobile IP networks to achieves seamless playback of continuous media streams even when both the mobile sender and receivers handoff simultaneously. Second, it analyzes the minimal buffer requirements of the mobile sender, the core router, the foreign agents and the mobile receivers in the multicast tree and formulates the initial playback delay within a handoff Guarantee Region (GR). Further, combined with the fine granularity scalability (FGS) encoding approach in the MPEG-4 standard, the SMM scheme achieves superior multimedia QoS guarantees and unlimited numbers of handoffs of the mobile sender and receivers only at the cost of degraded video quality for a short period after handoff with minimal extra bandwidth.

In this paper, we propose a set of broadcast streaming protocols designed for unidirectional radio channels. Considering the limited size and implementation overhead on a mobile terminal, the proposed protocol set is almost compliant with the current mobile streaming protocols, i.e. 3GPP PSS (Packet-switched Streaming Service), except for that the proposed protocols are designed to work on a unidirectional downlink channel. This protocol set enables flexible layout rendering by SMIL (Synchronized Multimedia Integration Language) in combination with SDP (Session Description Protocol), and reliable and synchronized static media (including still image and text) delivery by RTP (Real-time Transport Protocol) carousel. We present the prototype of this protocol set and measure its performance of video quality and waiting time for video presentation through a W-CDMA radio channel emulator and header compression nodes. From the experimental results, we show 1) trade-off between video quality and waiting time, 2) advantage and disadvantage of header compression, 3) effectiveness of synchronized transmission of SDP, SMIL, and I-frames of video objects, and 4) reliability of RTP-carousel. This protocol set is applicable to 3G MBMS (Multimedia Broadcast/Multicast Service) streaming service.

Point-to-point optical and millimeter wave communication has recently been of interest, especially in urban areas. Its benefits include simpler and easier installation compared with a land-based line. However, this technology suffers when adverse weather conditions are present, such as rain, fog and clouds, which induce scattering and absorption of the optical wave. The effects of scattering and absorption degrade the quality of the communication link resulting in increase of bit-error-rate. Therefore, there exists a need for accurate channel characterization in order to understand and mitigate the problem. In this paper, radiative transfer theory is employed to study the behavior of amplitude modulated signal propagating through a random medium. We show the effect of the medium to a modulated signal and relate the outcome on the quality of the communication link.

To provide multimedia services, a server system can use a cluster of servers to serve many users concurrently. In the literature, several assignment algorithms have been proposed to assign the media objects to these servers. The popularities of the media objects may change with time, and the service provider may add new objects or delete obsolete objects. In these cases, it is necessary to execute an assignment algorithm to find a new assignment. To realize the new assignment from the current one, it is necessary to relocate some objects. In this paper, we formulate and solve this relocation problem, and our objective is to minimize the relocation time for minimal service interruption. We propose a two-stage algorithm to solve this relocation problem. In the first stage, we introduce the concept of server relabelling, and apply the Hungarian method to find the largest common assignment between the new and current assignments. In the second stage, we propose a chain relocation schedule by which every object is directly relocated from its original server to the new server without deadlock. We use computer simulation to demonstrate the effectiveness of the optimal relocation method.