This paper presents an efficient and low-power-consumption parallel face-detection technology based on Haar-like features and implemented with a massive-parallel memory-embedded SIMD matrix. The massive-parallel memor...
详细信息
This paper presents an efficient and low-power-consumption parallel face-detection technology based on Haar-like features and implemented with a massive-parallel memory-embedded SIMD matrix. The massive-parallel memory-embedded SIMD matrix architecture has up to 2,048 2-bit processing elements, which are connected by a flexible switching network, and supports 2-bit 2,048-way bit-serial and word-parallel operations with a single command. For experimented verification of this matrix processing architecture, this parallel Haar-like-feature based face-detection technique has been implemented on an evaluation board and tested in practice. Evaluation results show that a total processing time of about 313 ms at 162 MHz clock frequency and 150 mW power dissipation can be realized. Thus, the reported parallel-face detection method with the massive-parallel memory-embedded SIMD matrix is a practical technology and is a promising solution for real-time mobile multimedia applications.
In designing realtime software, system designers need to find out the time budget to allocate to each action of realtime tasks so that the tasks can meet their deadlines. Our solution to this problem involves repres...
详细信息
In designing realtime software, system designers need to find out the time budget to allocate to each action of realtime tasks so that the tasks can meet their deadlines. Our solution to this problem involves representing the execution time of the actions as parameters, then analyzing the collaborative behavior of those realtime tasks. This paper proposes parametric timed models of realtime tasks whose executions are controlled by a scheduler. We develop an algorithm to synthesize a coherent model which represents the possible behavior from a set of realtime tasks by exhaustively searching their reachable states. A set of linear inequalities are then derived on the fly from the synthesized model as the condition of parameters for schedulability. By solving the inequalities using a constraint solver, we can obtain desirable values of the parameters. In addition, we have implemented the algorithm in a tool and conducted some experiments to show the effectiveness of our approach.
Videoconference systems have been around the market for a long time. Their aim is to provide a way of carrying out meetings without the need for having physical presence of the participants. However, the sense of real...
详细信息
Distributed Video Coding (DVC) is a popular topic in the research community and the past years have seen several different implementations. DVC has been proposed as a solution for applications that have limited batter...
详细信息
time and efficiency are matter of priority now. RFTD (Radio Frequency Tdentification) emerges as one of the converging technologies. While transportation plays an important role in urbanization, RFTD is now key cataly...
详细信息
Dynamic Voltage Scaling (DVS) has been widely used for decreasing the dynamic power dissipation of processors. For real-time systems, DVS techniques have been developed that permit to meet the timing constraints of mu...
详细信息
Dynamic Voltage Scaling (DVS) has been widely used for decreasing the dynamic power dissipation of processors. For real-time systems, DVS techniques have been developed that permit to meet the timing constraints of multiple real-time tasks and at the same time reduce the overall dynamic energy consumption. Known methods for static priority DVS scheduling are, however, either restricted to simple periodic/sporadic task release patterns or presume full a priori knowledge of task release times. Moreover, none of the present approaches considers the optimization of task priorities for reducing the energy consumption. In this paper we explore how to determine the static priorities and individual execution speeds (supply voltages) of multiple tasks with non-deterministic release times bounded by arrival curves such that the energy consumption is reduced and the real-time constraints are met. The result are different heuristics for the design of DVS-based real-time systems with static priorities. We show that the proposed methodology leads to energy-efficient system designs and demonstrate the applicability of the approach by means of experiments.
This live demonstration aims to show a real-time people tracking using an embedded smart stereo vision system (3DVS) consisting of two asynchronous (biologically-inspired) dynamic vision sensors and a processing unit ...
详细信息
This live demonstration aims to show a real-time people tracking using an embedded smart stereo vision system (3DVS) consisting of two asynchronous (biologically-inspired) dynamic vision sensors and a processing unit (DSP) embedding a detection and tracking algorithm.
The Web has become pervasive. This has led to a paradigm shift, where applications live on the Web as services, where they can be accessed using terminals of different types, such as regular computers, mobile devices,...
详细信息
The Web has become pervasive. This has led to a paradigm shift, where applications live on the Web as services, where they can be accessed using terminals of different types, such as regular computers, mobile devices, game consoles, and so on. The ability to dynamically combine data, code and other content from numerous web sites all over the world, and the ability to instantly publish services worldwide has opened up entirely new possibilities for software development. Such applications, referred to as mashups, are content aggregates that leverage the power of the Web to support instant, worldwide sharing of content. Unfortunately, for various reasons, the browser that is commonly used as the run-time environment of mashups is inadequate for hosting complex applications, in particular when considering embedded devices and subsystems that are not readily available in the web. In this paper, we introduce a runtime environment, intended for embedded devices, that is able to host mashups that combine data from the web and device specific peripherals, thus enabling context-aware mashups. As an example, we describe a mashup that combines map data from an existing map service and GPS data available from an external system.
In this paper, we proposed a blind watermarking technique to protect the video copyright and integrity by embedding both the robust and the fragile watermarks in H.264 video during video encoding. The proposed robust ...
详细信息
ISBN:
(纸本)9781424444618
In this paper, we proposed a blind watermarking technique to protect the video copyright and integrity by embedding both the robust and the fragile watermarks in H.264 video during video encoding. The proposed robust watermarks are embedded on some specific quantized DCT coefficients of intra-frame residue to keep the video copyright as those selected coefficients are strong enough to against the attack. On the other hand, our fragile watermarks are embedded on some motion vectors of inter-frames to preserve the video integrity with minimal affection to the video quality. The embedded messages are generated by Message-Digest algorithm 5 (MD5) to increase the detection sensitivity. Experimental results show that our proposed robust watermarking algorithm is robust to various attacks, and fragile watermarking algorithm is sensitive to the video forgery or malicious modification, such as transcoding and various signal processing. Additionally, the algorithm is viable for real-timeapplications.
暂无评论