Search

The objective of this paper is to develop an effective photographic style transfer method while preserving the semantic correspondence between the style and content images for both scenery and portrait images. A semantic correspondence guided photographic style transfer algorithm is developed, which is to ensure that the semantic structure of the content image has not been changed while the color of the style images is being migrated. The semantic correspondence is constructed in large scale regions based on image segmentation and also in local scale patches using Nearest-neighbor Field Search in the deep feature domain. Based on the semantic correspondence, a matting optimization is utilized to optimize the style transfer result to ensure the semantic accuracy and transfer faithfulness. The proposed style transfer method is further extended to automatically retrieve the style images from a database to make style transfer more-friendly. The experimental results show that our method could successfully conduct the style transfer while preserving semantic correspondence between diversity of scenes. A user study also shows that our method outperforms state-of-the-art photographic style transfer methods.

The dynamic optimization of virtual machine (VM) placement is to dynamically adjust the placement of VMs on physical machines (PMs) to accomplish some objectives with certain constraints. On one hand, the number of possible combinations of PMs and VMs can be extremely large, which make the optimal solution very hard to get. On the other hand, the optimized solution needs be reachable from the old solution. To solve the problem from both sides, a partitioned optimization framework is proposed. First, four different states of PM, i.e. off, sleeping, ready and running, are introduced with different energy consumption. Running pool, sleeping pool and off pool are set up which partition PMs based on their different states. The classification helps us build the energy consumption model which is needed to evaluate mapping solutions. To decide if a new solution is better than the old one, only three parts of energy need be considered, i.e. energy changes for PMs in different states, energy consumed for changing the states of PMs, and extra energy consumption for migrating VMs. An energy model composed of these three parts is built as the optimization objective. A method is presented to decide the most suitable range to conduct the energy optimization through excluding some PMs in the sleep or off pool if the best solution achieved with those PMs included cannot be better than old solution. A memetic algorithm combining the partheno-genetic algorithm with the multiplayer random evolutionary game theory is proposed to achieve the global optimal solution and generate the executable live migration sequence from old mapping to the target one at the same time. According to our experimental results, our method can decrease the optimization range remarkably. Within the optimized scales, the proposed algorithm performed very well to approach the global optimal solution and guarantee the solution’s feasibility from old solution at the same time.

Image inpainting is a challenging task due to the loss of the image information. Recently, GAN-based approaches have shown promising performance in the field of image inpainting. For this task, a superior similarity measurement of extracted patches from known and missing regions is important. Existing approaches usually adopt cosine distance to measure this similarity for missing region reconstruction. However, from the semantic-level perspective, these methods often generate content with inconsistent color and disorder structure due to the ignorance of the magnitude distance of the attended patches. To resolve this problem, we propose a Hybrid Contextual Attention Network (HCA-Net) with a novel attention module called hybrid contextual attention module (HCAM). HCAM takes account of both cosine distance and Euclidean distance as the measurement of the extracted patches and gives a better prediction of missing features. Besides, a Spectral-Normalization patch discriminator and the cosine loss are added into the model for patch-level and pixel-level consistency enhancement. Extensive results on three public datasets (Paris Street View, Celeba-HQ, and Places2), have both validated that our approach significantly outperforms the state-of-the-art approaches.

A Petri net based framework is proposed for automatic high level video event description, recognition and reasoning purposes. In comparison with the existing approaches reported in the literature, our work is characterized with a number of novel features: (i) the high level video event modeling and recognition based on Petri net are fully automatic, which are not only capable of covering single video events but also multiple ones without limit; (ii) more variations of event paths can be found and modeled using the proposed algorithms; (iii) the recognition results are more accurate based on automatic built high level event models. Experimental results show that the proposed method outperforms the existing benchmark in terms of recognition precision and recall. Additional advantages can be achieved such that hidden variations of events hardly identified by humans can also be recognized.

A Petri net based framework is proposed for automatic high level video event description, recognition and reasoning purposes. In comparison with the existing approaches reported in the literature, our work is characterized with a number of novel features: (i) the high level video event modeling and recognition based on Petri net are fully automatic, which are not only capable of covering single video events but also multiple ones without limit; (ii) more variations of event paths can be found and modeled using the proposed algorithms; (iii) the recognition results are more accurate based on automatic built high level event models. Experimental results show that the proposed method outperforms the existing benchmark in terms of recognition precision and recall. Additional advantages can be achieved such that hidden variations of events hardly identified by humans can also be recognized.

Carbapenem-resistant Enterobacteriaceae have been a global public health issue in recent years. Here, a carbapenem-resistant Kluyvera cryocrescens strain SCW13 was isolated from hospital sewage, and was then subjected to whole-genome sequencing (WGS). Based on WGS data, antimicrobial resistance genes were identified. Resistance plasmids were completely circularized and further bioinformatics analyses of plasmids were performed. A conjugation assay was performed to identify a self-transmissible plasmid mediating carbapenem resistance. A phylogenetic tree was constructed based on the core genome of publicly available Kluyvera strains. The isolate SCW13 exhibited resistance to cephalosporin and carbapenem. blaNDM-1 was found to be located on a ~53-kb self-transmissible IncX3 plasmid, which exhibited high similarity to the previously reported pNDM-HN380, which is an epidemic blaNDM-1-carrying IncX3 plasmid. Further, we found that SCW13 contained a chromosomal blaKLUC-2 gene, which was the probable origin of the plasmid-born blaKLUC-2 found in Enterobacter cloacae. Phylogenetic analysis showed that K. cryocrescens SCW13 exhibited a close relationship with K. cryocrescens NCTC10483. These findings highlight the further dissemination of blaNDM through clonal IncX3 plasmids related to pNDM-HN380 among uncommon Enterobacteriaceae strains, including Kluyvera in this case.

Carbapenem-resistant Enterobacteriaceae have been a global public health issue in recent years. Here, a carbapenem-resistant Kluyvera cryocrescens strain SCW13 was isolated from hospital sewage, and was then subjected to whole-genome sequencing (WGS). Based on WGS data, antimicrobial resistance genes were identified. Resistance plasmids were completely circularized and further bioinformatics analyses of plasmids were performed. A conjugation assay was performed to identify a self-transmissible plasmid mediating carbapenem resistance. A phylogenetic tree was constructed based on the core genome of publicly available Kluyvera strains. The isolate SCW13 exhibited resistance to cephalosporin and carbapenem. blaNDM-1 was found to be located on a ~53-kb self-transmissible IncX3 plasmid, which exhibited high similarity to the previously reported pNDM-HN380, which is an epidemic blaNDM-1-carrying IncX3 plasmid. Further, we found that SCW13 contained a chromosomal blaKLUC-2 gene, which was the probable origin of the plasmid-born blaKLUC-2 found in Enterobacter cloacae. Phylogenetic analysis showed that K. cryocrescens SCW13 exhibited a close relationship with K. cryocrescens NCTC10483. These findings highlight the further dissemination of blaNDM through clonal IncX3 plasmids related to pNDM-HN380 among uncommon Enterobacteriaceae strains, including Kluyvera in this case.

Recent developments of video super-resolution reconstruction often exploit spatial and temporal contexts from input frame sequence by making use of explicit motion estimation, e.g., optical flow, which may introduce accumulated errors and requires huge computations to obtain an accurate estimation. In this paper, we propose a novel multi-branch dilated convolution module for real-time frame alignment without explicit motion estimation, which is incorporated with the depthwise separable up-sampling module to formulate a sophisticated real-time video super-resolution network. Specifically, the proposed video super-resolution framework can efficiently acquire a larger receptive field and learn spatial–temporal features of multiple scales with minimal computational operations and memory requirements. Extensive experiments show that the proposed super-resolution network outperforms current state-of-the-art real-time video super-resolution networks, e.g., VESPCN and 3DVSRnet, in terms of PSNR values (0.49 dB and 0.17 dB) on average in various datasets, but requires less multiplication operations.

In recent years, convolutional neural networks (CNNs) have accelerated the developments of video super resolution (SR) for achieving higher image quality. However, the computational cost of existing CNN-based video super-resolution is too heavy for real-time applications. In this paper, we propose a new video super-resolution framework using lightweight frame alignment module and well-designed up-sampling module for real-time processing. Specifically, our framework, which is called as Lightweight Shuffle Video Super-Resolution Network (LSVSR), combines channel shuffling, depthwise convolution and pointwise group convolution to significantly reduce the computational burden during frame alignment and high-resolution frame reconstruction. On the public benchmark datasets, our proposed network outperforms the state-of-the-art lightweight video SR networks in terms of objective (PSNR and SSIM) and subjective evaluations, number of network parameters and floating-point operations. Our network can achieve real-time 540P to 2160P 4 × super-resolution for more than 60fps using desktop GPUs or mobile phones with neural processing unit.

Contend-based flower image retrieval is one of the hottest and most challenging problem in content-based image retrieval area. In this paper, a flower image retrieval method is proposed based on a memetic algorithm. The proposed method, which combines a global search strategy with a local search strategy, uses a memetic algorithm to select the optimal feature subset. Genetic algorithm is used as the global search strategy while approximate Markov blanket is adopted as the local search strategy. Primary classifiers are trained using the proposed memetic algorithm for each kind of features. The probabilities obtained by all the primary classifiers are combined together to form a mid-level feature used to train the final classifier. Experimental results show that the proposed method selects fewer number of features with better precisions and recall ratios. That also brings improvements on retrieval time.

The computational model of energy consumption is built to serve as an evaluation function.An algorithm based on evolutionary game theory is proposed to solve the problem of dynamic VMs placement.It is analyzed that the algorithm can theoretically reach the optimal solution of the dynamic VMs placement problem.The algorithm can take the initial mapping into account and generate an executable list of VMs live migrations from the initial state to the target state. Power saving of data centers has become an urgent problem in recent years. For a virtualized data center, optimizing the placement of virtual machines (VMs) dynamically is one of the most effective methods for power savings. Based on a deep study on VMs placement, a solution is proposed and described in this paper to solve the problem of dynamic placement of VMs toward optimization of their energy consumptions. A computational model of energy consumption is proposed and built. A novel algorithm based on evolutionary game theory is also presented, which successfully addresses the challenges faced by dynamic placement of VMs. It is proved that the proposed algorithm can reach the optimal solutions theoretically. Experimental results also demonstrate that, by adjusting VMs placement dynamically, the energy consumption can be reduced correspondingly. In comparison with the existing state of the arts, our proposed method outperforms other five algorithms tested and achieves savings of 30-40% on energy consumption.

The objective of this paper is to develop an effective photographic transfer method while preserving the semantic correspondence between the style and content images. A semantic correspondence guided deep photo style transfer algorithm is developed, which is to ensure that the semantic structure of the content image has not been changed while the color of the style images is being migrated. The semantic correspondence is constructed in large scale regions based on image segmentation and also in local scale patches using deep image analogy. Based on the semantic correspondence, a matting optimization is utilized to optimize the style transfer result to ensure the semantic accuracy and transfer faithfulness. The proposed style transfer method is further extended to automatically retrieve the style images from a database to make style transfer more-friendly. The experimental results show that our method could successfully conduct the style transfer while preserving semantic correspondence between diversity of scenes. A user study also shows that our method outperforms state-of-the-art photographic style transfer methods.

Automatic fine-art painting classification is an important task to assist the analysis of fine-art paintings. In this paper, we propose a novel two-channel deep residual network to classify fine-art painting images. In detail, we take the advantage of the ImageNet to pre-train the deep residual network. Our two channels include the RGB channel and the brush stroke information channel. The gray-level co-occurrence matrix is used to detect the brush stroke information, which has never been considered in the task of fine-art painting classification. Experiments demonstrate that the proposed model achieves better classification performance than other models. Moreover, each stage of our model is effective for the image classification.

As the task load of cloud system grows bigger, it becomes very important to design an efficiency task scheduling algorithm. This paper proposes a task scheduling algorithm based on genetic algorithm and ant colony optimization algorithm. The hybrid task scheduling algorithm can help the cloud system to complete users’ tasks faster. Simulation experiment results in CloudSim show that, comparing with genetic algorithm and ant colony optimization algorithm alone, the hybrid algorithm has better performance in the aspects of load balancing and optimal time span.

Effective methods of workflow scheduling can improve the performance of workflow systems. Based on the study of existing scheduling methods, a method of workflow scheduling, called phased method, is proposed. This method is based on colored Petri nets. Activities of workflows are divided into several groups to be scheduled in different phases using this method. Details of the method are discussed. Experimental results show that the proposed method can deal with the uncertainties and the dynamic circumstances very well and a satisfactory balance can be achieved between static global optimization and dynamic local optimization.

The large-scale video data on the web contain a lot of semantics, which are an important part of semantic web. Video descriptors can usually represent somewhat the semantics. Thus, they play a very important role in web multimedia content analysis, such as Scale-invariant feature transform SIFT feature. In this paper, we proposed a new video descriptor, called a temporal-compress and shorter SIFTTC-S-SIFT which can efficiently and effectively represent the semantics of web videos. By omitting the least discriminability orientation in three stages of standard SIFT on every representative frame, the dimensions of the shorter SIFT are reduced from 128-dimension to 96-dimension to save space storage. Then, the SIFT can be compressed by tracing SIFT features on video temporal domain, which highly compress the quantity of local features to reduce visual redundancy, and keep basically the robustness and discrimination. Experimental results show our method can yield comparable accuracy and compact storage size.

Inverted index is a key component in a search engine. Due to the high update frequency of web pages, incremental inverted index is commonly used in modern web-based search engine. In order to improve the performance of search engine, an efficient data block and control block DB&CB structure for incremental inverted index is proposed. Theoretical analysis and experiment results show that the average complexities of posting lists' append, insert and delete operations are O(1), the average inverted index construction time is reduced by 37% comparing with the linked block data structure and the query processing performance is as good as others.

Pre-scheduling of workflows is to schedule all tasks of a workflow instance as soon as it is initialized. Based on the study of existing scheduling methods, a new optimization method of workflow pre-scheduling is proposed based on Nested Genetic Algorithm(NGA). The solutions of resource allocation and execution sequence are optimized and global optimization for a single process instance can be approached using the proposed method. Experiments are done to show the feasibility and superiority of the proposed method.

Using effective modeling ways can improve the quality of workflow models. A new role-oriented modeling way is presented to set up workflow process models based on object Petri net. The details of each role are encapsulated in a role object net. The interfaces are provided for it to communicate with other roles. The cooperative relationships of roles are described in control object nets. An example is given to show the process of the presented modeling method.

Accurate performance metric models are the key to Web capacity planning related problems. Due to the complexity of Web systems, analytical modeling without integrating the performance testing process is not enough to get accurate metric models. To integrate performance testing and analytical modeling in a systematic way, a Web performance modeling process is presented based on the methodology of learning from data. The process divides the modeling activity into several phases: constructing models and hypothetical conditions, deriving test cases, estimating parameters and validating models, etc. The scalability of a real Web community system (www.igroot.com) is studied by using the proposed process. The error of estimated saturation point is within 1 percent, the error of estimated lower bound of buckle point is within 5 percent. At last, a HTTP processing bottleneck at the architecture level is identified by correlating the model with the threads data of the Web server.

A resource allocation method is proposed to determine the proper number of resources added to each resource class with cost constraint in order to optimize workflow time performance. The average throughput time of workflow instances is used to measure the workflow time performance. An approach which calculates the average throughput time of workflow instances is proposed. An improved genetic algorithm is presented to realize the allocation method. Experimental results show that the algorithm has good evolution performance and is superior to other allocation methods.

How to allocate workflow resources with cost constraint to optimize workflow time performance was studied. The average throughput time of workflow instances is used to measure the workflow time performance. A method based on queuing theory was proposed to calculate the average throughput time of workflow instances. Since the problem can be come down to an unbounded knapsack problem (UKP), an improved GA (Genetic Algorithm) suitable to solve the UKP was proposed to solve the problem. Examples were given to illustrate the feasibility and validity of the method. The results show that the algorithm has good evolution performance and can achieve or approach the optimal solutions. And compared with other allocation methods, our method performs best.

Workflow resource model describes all kinds of resources that support the execution of workflows. The meta-model for workflow resource model presents the constituents of workflow resource model. It is one of the three correlative sub-meta-models for workflow model. Based on the analysis of existed studies and real cases, an extended meta-model for workflow resource model was introduced by extending and modifying the meta-model for organizational model proposed by WfMC. The detail of entities and their relationships were described. The relationships between workflow resource model and process model were discussed. XML was used to describe the meta-model. In the end, a conclusion and proposals for future research directions were presented.

The arm of this work is to provide a formal model and an effective way for information retrieving from a big related data set. Based upon fuzzy logic operation, a fuzzy mathematical model of 0-1 mixture programming is addressed. Meanwhile, a density function indicating the overall possessive status of the effective mined out data is introduced. Then, a soft computing (SC) approach which is a genetic algorithm (GA) embedded fuzzy deduction is presented. During the SC process, fuzzy logic decision is taken into the uses of determining the genes' length, calculating fitness function and choosing feasible solution. Stimulated experiments and comparison tests show that the methods can match the user's most desired information from magnanimity data exactly and efficiently. The approaches can be extended in practical application in solving general web mining problem.

The total amount of resources in a workflow process is constrained by the cost of employment; meanwhile, the assignment of resources directly decides the average waiting and dealing time of each activity in a workflow. By optimizing the allocation of resources, it can minimize the total flow time of a workflow. Some optimization rules for minimizing the average responding time of a workflow are addressed in this paper. Then, a mathematical model of resources optimization with minimal cost time is designed, and an improved Genetic Algorithm to resolve this optimization problem is presented. The efficiency of the approaches presented is evaluated by simulated experiments and comparison with tradition GA.