Brian Foo

Optimality and improvement of dynamic voltage scaling algorithms for multimedia applications

The high complexity and time-varying workload of emerging multimedia applications poses a major challenge for dynamic voltage scaling (DVS) algorithms. Although many DVS algorithms have been proposed for real-time applications, an efficient method for evaluating the optimality of such DVS algorithms for multimedia applications does not yet exist. In this paper, we propose the first offline linear programming (LP) method to determine the minimum energy consumption for processing multimedia tasks under stringent delay deadlines. On the basis of the obtained energy lower bound, we evaluate the optimality of various existing DVS algorithms. Furthermore, we extend the LP formulation in order to construct an online DVS algorithm for real-time multimedia processing based on robust sequential linear programming. Simulation results obtained by decoding a wide range of video sequences show that, on average, our online algorithm provides a scheduling solution that requires less than 0.3% more energy than the optimal lower bound with only 0.03% miss rate. In comparison, a very recent algorithm consumes approximately 4% more energy than the optimal lower bound at the same miss rate.

Analytical Rate-Distortion-Complexity Modeling of Wavelet-Based Video Coders

Analytical modeling of the performance of video coders is essential in a variety of applications, such as power-constrained processing, complexity-driven video streaming, etc., where information concerning rate, distortion, or complexity (and their interrelation) is required. In this paper, we present a novel rate-distortion-complexity (R-D-C) analysis for state-of-the-art wavelet video coding methods by explicitly modeling several aspects found in operational coders, i.e., embedded quantization, quadtree decompositions of block significance maps and context-adaptive entropy coding of subband blocks. This paper achieves two main goals. First, unlike existing R-D models for wavelet video coders, the proposed derivations reveal for the first time the expected coding behavior of specific coding algorithms (e.g., quadtree decompositions, coefficient significance, and refinement coding) and, therefore, can be used for a variety of coding mechanisms incorporating some or all the coding algorithms discussed. Second, the proposed modeling derives for the first time analytical estimates of the expected number of operations (complexity) of a broad class of wavelet video coding algorithms based on stochastic source models, the coding algorithm characteristics and the system parameters. This enables the formulation of an analytical model characterizing the complexity of various video decoding operations. As a result, this paper complements prior complexity-prediction research that is based on operational measurements. The accuracy of the proposed analytical R-D-C expressions is justified against experimental data obtained with a state-of-the-art motion-compensated temporal filtering based wavelet video coder, and several new insights are revealed on the different tradeoffs between rate-distortion performance and the required decoding complexity.

A Distributed Approach for Optimizing Cascaded Classifier Topologies in Real-Time Stream Mining Systems

In this paper, we discuss distributed optimization techniques for configuring classifiers in a real-time, informationally-distributed stream mining system. Due to the large volume of streaming data, stream mining systems must often cope with overload, which can lead to poor performance and intolerable processing delay for real-time applications. Furthermore, optimizing over an entire system of classifiers is a difficult task since changing the filtering process at one classifier can impact both the feature values of data arriving at classifiers further downstream and, thus, the classification performance achieved by an ensemble of classifiers, as well as the end-to-end processing delay. To address this problem, this paper makes three main contributions. 1) Based upon classification and queuing theoretic models, we propose a utility metric that captures both the performance and the delay of a binary filtering classifier system. 2) We introduce a low-complexity framework for estimating the system utility by observing, estimating, and/or exchanging parameters between the interrelated classifiers deployed across the system. 3) We provide distributed algorithms to reconfigure the system, and analyze the algorithms based upon their convergence properties, optimality, information exchange overhead, and rate of adaptation to nonstationary data sources. We provide results using different video classifier systems.

Optimality and Improvement of Dynamic Voltage Scaling Algorithms for Multimedia Applications

The high complexity and time-varying workload of emerging multimedia applications poses a major challenge for dynamic voltage scaling (DVS) algorithms. Although many DVS algorithms have been proposed for real-time applications, an efficient method for evaluating the optimality of such DVS algorithms for multimedia applications does not yet exist. In this paper, we propose the first offline linear programming (LP) method to determine the minimum energy consumption for processing multimedia tasks under stringent delay deadlines. On the basis of the obtained energy lower bound, we evaluate the optimality of various existing DVS algorithms. Furthermore, we extend the LP formulation in order to construct an online DVS algorithm for real-time multimedia processing based on robust sequential linear programming. Simulation results obtained by decoding a wide range of video sequences show that, on average, our online algorithm provides a scheduling solution that requires less than 0.3% more energy than the optimal lower bound with only 0.03% miss rate. In comparison, a very recent algorithm consumes approximately 4% more energy than the optimal lower bound at the same miss rate.

Configuring topologies of distributed semantic concept classifiers for continuous multimedia stream processing

Real-time multimedia semantic concept detection requires instant identification of a set of concepts in streaming video or images. However, the potentially high data volumes of multimedia content, and high complexity associated with individual concept detectors, have hindered its practical deployment. In this paper, we present a new online concept detection system deployed on top of a distributed stream mining system. It uses a tree-topology of classifiers that are constructed on a semantic hierarchy of concepts of interest. We introduce a novel methodology for configuring such cascaded classifier topologies under constraints on the available resources. In our approach, we configure individual classifiers with optimized operating points after jointly and explicitly considering the misclassification cost of each end-to-end class of interest in the tree, the system imposed resource constraints, and the confidence level of each object that is classified. We describe the implemented application, system, and optimization algorithms, and verify that significant improvement in terms of accuracy of classification can be achieved through our approach.

A Queuing Theoretic Approach to Processor Power Adaptation for Video Decoding Systems

Video decoding applications must often cope with highly time-varying workload demands, while meeting stringent display deadlines. Voltage/frequency scalable processors are highly attractive for video decoding on resource-constrained systems, since significant energy savings can be achieved by dynamically adapting the processor speed based on the changing workload demand. Previous works on video-related voltage scaling algorithms are often limited by the lack of a good complexity model for video and often do not explicitly consider the video quality impact of various steps involved in the decoding process. Our contribution in this paper is threefold. First, we propose a novel complexity model through offline training that explicitly considers the video source characteristics, the encoding algorithm, and platform specifics to predict execution times. Second, based on the complexity model, we propose low complexity online voltage scaling algorithms to process decoding jobs such that they meet their display deadlines with high probability. We show that on average, our queuing-based voltage scaling algorithm provides approximately 10%-15% energy savings over existing voltage scaling algorithms. Finally, we propose a joint voltage scaling and quality-aware priority scheduling algorithm that decodes jobs in order of their distortion impact, such that by setting the processor to various power levels and decoding only the jobs that contribute most to the overall quality, efficient quality, and energy tradeoffs can be achieved. We demonstrate the scalability of our algorithm in various practical decoding scenarios, where reducing the power to 25% of the original power can lead to quality degradations of less than 1.0 dB PSNR.

Distributed classifier chain optimization for real-time multimedia stream mining systems

We consider the problem of optimally configuring classifier chains for real-time multimedia stream mining systems. Jointly maximizing the performance over several classifiers under minimal end-to-end processing delay is a difficult task due to the distributed nature of analytics (e.g. utilized models or stored data sets), where changing the filtering process at a single classifier can have an unpredictable effect on both the feature values of data arriving at classifiers further downstream, as well as the end-to-end processing delay. While the utility function can not be accurately modeled, in this paper we propose a randomized distributed algorithm that guarantees almost sure convergence to the optimal solution. We also provide results using speech data showing that the algorithm can perform well under highly dynamic environments.

Configuring Trees of Classifiers in Distributed Multimedia Stream Mining Systems

Multimedia stream mining applications require the identification of several different attributes in data content, and hence rely on a set of cascaded statistical classifiers to filter and process the data dynamically. In this paper, we introduce a novel methodology for configuring such cascaded classifier topologies, specifically binary classifier trees, in resource-constrained, distributed stream mining systems. Instead of traditional load shedding, our approach configures classifiers with optimized operating points after jointly considering the misclassification cost of each end-to-end class of interest in the tree, the resource constraints for every classifier, and the confidence level of each data object that is classified. The proposed approach allows for both intelligent load shedding as well as data replication based on available resources dynamically. We evaluate the algorithm on a sports video concept detection application and identify huge cost savings over load shedding alone. Additionally, we propose several distributed algorithms that enable each classifier in the tree to reconfigure itself based on local information exchange. We analyze the associated tradeoffs between convergence time, information overhead, and the cost efficiency of results achieved by each classifier for each of these algorithms.

A rules-based approach for configuring chains of classifiers in real-time stream mining systems

Networks of classifiers can offer improved accuracy and scalability over single classifiers by utilizing distributed processing resources and analytics. However, they also pose a unique combination of challenges. First, classifiers may be located across different sites that are willing to cooperate to provide services, but are unwilling to reveal proprietary information about their analytics, or are unable to exchange their analytics due to the high transmission overheads involved. Furthermore, processing of voluminous stream data across sites often requires load shedding approaches, which can lead to suboptimal classification performance. Finally, real stream mining systems often exhibit dynamic behavior and thus necessitate frequent reconfiguration of classifier elements to ensure acceptable end-to-end performance and delay under resource constraints. Under such informational constraints, resource constraints, and unpredictable dynamics, utilizing a single, fixed algorithm for reconfiguring classifiers can often lead to poor performance. In this paper, we propose a new optimization framework aimed at developing rules for choosing algorithms to reconfigure the classifier system under such conditions. We provide an adaptive, Markov model-based solution for learning the optimal rule when stream dynamics are initially unknown. Furthermore, we discuss how rules can be decomposed across multiple sites and propose a method for evolving new rules from a set of existing rules. Simulation results are presented for a speech classification system to highlight the advantages of using the rules-based framework to cope with stream dynamics.

Analytical Complexity Modeling of Wavelet-based Video Coders

Analytical modeling for video coders can be used in a variety of scenarios where information concerning rate, distortion or complexity is essential for driving system or network interactions with media algorithms. While rate and distortion modeling have been covered extensively in previous works, complexity is not well addressed because it is highly algorithm dependent and hence difficult to model. Based on a stochastic modeling framework for the transform coefficients, we present a novel complexity analysis for state-of-the-art wavelet video coding methods by explicitly modeling several aspects found in operational coders, i.e. embedded quantization and quadtree decompositions of block significance maps. The proposed modeling derives for the first time analytical estimates of the expected number of operations (complexity) for a broad class of wavelet video coders based on stochastic source models, coding algorithm and system parameters.