Search

Seasonal demand forecasting is critical for effective supply chain management. However, conventional forecasting methods face difficulties accurately estimating seasonal variations, owing to time-varying demand trends and limited data availability. In this paper, we propose a Fourier time-varying grey model (FTGM) to tackle this issue. The FTGM builds upon grey models, which are effective with limited data, and leverages Fourier functions to approximate time-varying parameters that allow it to represent seasonal variations. A data-driven selection algorithm adaptively determines the appropriate Fourier order of the FTGM without prior knowledge of data characteristics. Using the well-known M5 competition data, we compare our model with state-of-the-art forecasting methods taken from grey models, statistical methods, and architectures of neural network-based methods. The experimental results show that the FTGM outperforms popular seasonal forecasting methods in terms of standard accuracy metrics, providing a competitive alternative for seasonal demand forecasting in retail companies.

Urged by a close future perspective of a traffic flow made of a mix of human-driven vehicles and automated vehicles (AVs), research has recently focused on studying the traffic flow characteristics of Adaptive Cruise Controls (ACCs), the most typical AV. However, in most works, the ACC system is studied under a simplifying and unrealistic assumption, or the ACC system modeled is inaccurate. This paper proposes a general hierarchical control system to model ACC systems with several assumptions based on the deficiencies above. Moreover, a field experiment was conducted, and the corresponding experimental data was used to verify the proposed hierarchical control system and assumptions. In addition, string stability is explored along with sensitivity analyses of control parameters based on an example under the constant time gap policy. The results show that different upper-level controller parameters have different delays, where the delay of the speed is negligible; the introduction of actuator delay and lag in the lower-level controller can significantly improve the model goodness of fit. Furthermore, optimizing the delay and lag in the lower-level controller can significantly enhance the string stability of ACCs than optimizing the control parameters., Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public., Electronic Instrumentation

Contains fulltext : 246032.pdf (Publisher’s version ) (Open Access), Deep neural networks (DNNs) are an indispensable machine learning tool despite the difficulty of diagnosing what aspects of a model's input drive its decisions. In countless real-world domains, from legislation and law enforcement to healthcare, such diagnosis is essential to ensure that DNN decisions are driven by aspects appropriate in the context of its use. The development of methods and studies enabling the explanation of a DNN’s decisions has thus blossomed into an active and broad area of research. The field's complexity is exacerbated by competing definitions of what it means "to explain" the actions of a DNN and to evaluate an approach's "ability to explain". This article offers a field guide to explore the space of explainable deep learning for those in the AI/ML field who are uninitiated. The field guide: i) Introduces three simple dimensions defining the space of foundational methods that contribute to explainable deep learning, ii) discusses the evaluations for model explanations, iii) places explainability in the context of other related deep learning research areas, and iv) discusses user-oriented explanation design and future directions. We hope the guide is seen as a starting point for those embarking on this research field.

Contains fulltext : 246032.pdf (Publisher’s version ) (Open Access), Deep neural networks (DNNs) are an indispensable machine learning tool despite the difficulty of diagnosing what aspects of a model's input drive its decisions. In countless real-world domains, from legislation and law enforcement to healthcare, such diagnosis is essential to ensure that DNN decisions are driven by aspects appropriate in the context of its use. The development of methods and studies enabling the explanation of a DNN’s decisions has thus blossomed into an active and broad area of research. The field's complexity is exacerbated by competing definitions of what it means "to explain" the actions of a DNN and to evaluate an approach's "ability to explain". This article offers a field guide to explore the space of explainable deep learning for those in the AI/ML field who are uninitiated. The field guide: i) Introduces three simple dimensions defining the space of foundational methods that contribute to explainable deep learning, ii) discusses the evaluations for model explanations, iii) places explainability in the context of other related deep learning research areas, and iv) discusses user-oriented explanation design and future directions. We hope the guide is seen as a starting point for those embarking on this research field.

© 2020 Elsevier Ltd This study focuses on Brazil, Russia, India, China and South Africa (BRICS countries), which contribute over 40% of global CO2 emissions. Using panel co-integration tests, fully modified OLS and seemingly unrelated regressions, the study contributes to the literature by revealing that public debt securities foster the transition from fossil fuel electricity towards low-carbon electricity, whereas private credit is mostly profitless for electricity production transition. The explanation is that environmental pressure urges public capital to play a vital role in electricity transition, while bank loans are reluctant to leave the electricity from fossil fuel for considerable returns. The installed capacity of electricity stations drives the association between financial capital and electricity production. Financial markets in China and South Africa play a more significant role in electricity transition than the other countries. Low-carbon electricity transition requires transformation of financial markets in all these countries.

© 2020 Elsevier Ltd This study focuses on Brazil, Russia, India, China and South Africa (BRICS countries), which contribute over 40% of global CO2 emissions. Using panel co-integration tests, fully modified OLS and seemingly unrelated regressions, the study contributes to the literature by revealing that public debt securities foster the transition from fossil fuel electricity towards low-carbon electricity, whereas private credit is mostly profitless for electricity production transition. The explanation is that environmental pressure urges public capital to play a vital role in electricity transition, while bank loans are reluctant to leave the electricity from fossil fuel for considerable returns. The installed capacity of electricity stations drives the association between financial capital and electricity production. Financial markets in China and South Africa play a more significant role in electricity transition than the other countries. Low-carbon electricity transition requires transformation of financial markets in all these countries.

© 2020 Elsevier B.V. Multidimensional resource allocation is a hot topic in cloud computing, but current cloud platforms support only fixed resource allocation, that is, the user resource requirements are consistent throughout the usage period, which may cause a waste of resources and reduce the revenue of resource providers. Therefore, time-varying multidimensional resource allocation and the corresponding pricing mechanism represent a new challenge in cloud computing. We address the problem of online time-varying multidimensional resource allocation and pricing in clouds. Specifically, (1) we propose a novel integer programming model for the time-varying multidimensional resource allocation problem and (2) we design a truthful online auction mechanism for resource allocation in a competitive environment. For the resource allocation algorithm, we propose a waiting period strategy and dominant-resource-based strategy to improve the social welfare and resource utilization. Simultaneously, a payment pricing algorithm based on critical value theory is proposed. Finally, we prove that the mechanism is truthful and individual rationality. Compared with existing research, our approach is characterized by high social welfare, high resource utilization and short execution time.

© 2020 the Partner Organisations. The most frequently used approach to preparing aggregation-induced emission fluorogen (AIEgen) particles is precipitation. Therefore, the addition of an AIEgen solution into water results in the formation of AIEgen particles in a very short time. Within such a short period of time and in the absence of proper mixing under shear, AIE particles are likely to be distributed in a wide range of sizes, thereby affecting their ultimate brightness and applications. Despite numerous attempts, the size of AIEgen particles is still within the range of 200-300 nm. For the first time, we developed a facile robust and cost-effective method for the fabrication of aggregation-induced emission nanoparticles with tuneable particle sizes <100 nm, high quantum yield, and excellent photostability. The direct diffusion of nanoparticles within the cell or in a single-celled organism, as an advantage of size reduction, opens new opportunities for biological and material studies. Such a significant reduction in AIE nanoparticle size has the potential for developing more efficient techniques for characterizing advanced nanomaterials and understanding biological processes and detection strategies.

The effect of nanosilica on the sulfate attack resistivity of cement mortar was investigated through study on the mechanical property evolution and the length change of the cement mortar under 5 wt.% sodium sulfate for 6 months. Meanwhile, the effects were compared with those of fly ash-replacement mortar. Results showed that by taking the advantages of nanosilica and fly ash in improving the property of cement mortar at early and later ages, the sulfate attack resistance of cement mortar can be enhanced in mechanical property increase and expansion reduction. Further, it implies that a combination of both pozzolans could enhance the sulfate attack resistivity of cement-based materials.

Crystallization is a conventional technique for fabricating physically cross-linked poly(vinyl alcohol)-based hydrogels (PVA) with desired properties. The creation of crystalline regions induced by the freeze-thaw process is attributed to the molecular organization that serves as the junction points containing folded chains. The current study employs an aggregation-induced emission fluorogen (AIEgen) to produce a new approach for studying the kinetics of the formation of crystalline regions induced by the freeze-thaw process. For the first time, this research directly links the effects of freeze-thaw cycles to the final physical and mechanical properties of PVA hydrogels by using AIEgen. The results of the study reveal an increase in the fluorescent properties during the freeze-thaw process that reflects the rise in the percentage of crystallinity. This approach can be employed effectively to demonstrate the crystallization process using significant color and brightness differences in crystalline regions compared to amorphous areas. This capability allows researchers to visualize the crystalline regions clearly in situ and to study their kinetics, formation, and transition in the solid state. The results of this study have implications for the broad class of gels derived through polymer crystallization from solution.

© 1989-2012 IEEE. Predicting the popularity of Point of Interest (POI) has become increasingly crucial for location-based services, such as POI recommendation. Most of the existing methods can seldom achieve satisfactory performance due to the scarcity of POI's information, which tendentiously confines the recommendation to popular scene spots, and ignores the unpopular attractions with potentially precious values. In this paper, we propose a novel approach, termed Hierarchical Multi-Clue Fusion (HMCF), for predicting the popularity of POIs. Specifically, in order to cope with the problem of data sparsity, we propose to comprehensively describe POI using various types of user generated content (UGC) (e.g., text and image) from multiple sources. Then, we devise an effective POI modelling method in a hierarchical manner, which simultaneously injects semantic knowledge as well as multi-clue representative power into POIs. For evaluation, we construct a multi-source POI dataset by collecting all the textual and visual content of several specific provinces in China from four main-stream tourism platforms during 2006 to 2017. Extensive experimental results show that the proposed method can significantly improve the performance of predicting the attractions' popularity as compared to several baseline methods.

© 1992-2012 IEEE. In this paper, we study the problem of cross-modal retrieval by hashing-based approximate nearest neighbor search techniques. Most existing cross-modal hashing works mainly address the issue of multi-modal integration complexity using the same mapping and similarity calculation for data from different media types. Nonetheless, this may cause information loss during the mapping process due to overlooking the specifics of each individual modality. In this paper, we propose a simple yet effective cross-modal hashing approach, termed collective reconstructive embeddings (CRE), which can simultaneously solve the heterogeneity and integration complexity of multi-modal data. To address the heterogeneity challenge, we propose to process heterogeneous types of data using different modality-specific models. Specifically, we model textual data with cosine similarity-based reconstructive embedding to alleviate the data sparsity to the greatest extent, while for image data, we utilize the Euclidean distance to characterize the relationships of the projected hash codes. Meanwhile, we unify the projections of text and image to the Hamming space into a common reconstructive embedding through rigid mathematical reformulation, which not only reduces the optimization complexity significantly but also facilitates the inter-modal similarity preservation among different modalities. We further incorporate the code balance and uncorrelation criteria into the problem and devise an efficient iterative algorithm for optimization. Comprehensive experiments on four widely used multimodal benchmarks show that the proposed CRE can achieve a superior performance compared with the state of the art on several challenging cross-modal tasks.

© 2013 IEEE. Video captioning has been attracting broad research attention in the multimedia community. However, most existing approaches heavily rely on static visual information or partially capture the local temporal knowledge (e.g., within 16 frames), thus hardly describing motions accurately from a global view. In this paper, we propose a novel video captioning framework, which integrates bidirectional long-short term memory (BiLSTM) and a soft attention mechanism to generate better global representations for videos as well as enhance the recognition of lasting motions in videos. To generate video captions, we exploit another long-short term memory as a decoder to fully explore global contextual information. The benefits of our proposed method are two fold: 1) the BiLSTM structure comprehensively preserves global temporal and visual information and 2) the soft attention mechanism enables a language decoder to recognize and focus on principle targets from the complex content. We verify the effectiveness of our proposed video captioning framework on two widely used benchmarks, that is, microsoft video description corpus and MSR-video to text, and the experimental results demonstrate the superiority of the proposed approach compared to several state-of-the-art methods.

© 1989-2012 IEEE. Predicting the popularity of Point of Interest (POI) has become increasingly crucial for location-based services, such as POI recommendation. Most of the existing methods can seldom achieve satisfactory performance due to the scarcity of POI's information, which tendentiously confines the recommendation to popular scene spots, and ignores the unpopular attractions with potentially precious values. In this paper, we propose a novel approach, termed Hierarchical Multi-Clue Fusion (HMCF), for predicting the popularity of POIs. Specifically, in order to cope with the problem of data sparsity, we propose to comprehensively describe POI using various types of user generated content (UGC) (e.g., text and image) from multiple sources. Then, we devise an effective POI modelling method in a hierarchical manner, which simultaneously injects semantic knowledge as well as multi-clue representative power into POIs. For evaluation, we construct a multi-source POI dataset by collecting all the textual and visual content of several specific provinces in China from four main-stream tourism platforms during 2006 to 2017. Extensive experimental results show that the proposed method can significantly improve the performance of predicting the attractions' popularity as compared to several baseline methods.

© 2013 IEEE. Video captioning has been attracting broad research attention in the multimedia community. However, most existing approaches heavily rely on static visual information or partially capture the local temporal knowledge (e.g., within 16 frames), thus hardly describing motions accurately from a global view. In this paper, we propose a novel video captioning framework, which integrates bidirectional long-short term memory (BiLSTM) and a soft attention mechanism to generate better global representations for videos as well as enhance the recognition of lasting motions in videos. To generate video captions, we exploit another long-short term memory as a decoder to fully explore global contextual information. The benefits of our proposed method are two fold: 1) the BiLSTM structure comprehensively preserves global temporal and visual information and 2) the soft attention mechanism enables a language decoder to recognize and focus on principle targets from the complex content. We verify the effectiveness of our proposed video captioning framework on two widely used benchmarks, that is, microsoft video description corpus and MSR-video to text, and the experimental results demonstrate the superiority of the proposed approach compared to several state-of-the-art methods.

© 1992-2012 IEEE. In this paper, we study the problem of cross-modal retrieval by hashing-based approximate nearest neighbor search techniques. Most existing cross-modal hashing works mainly address the issue of multi-modal integration complexity using the same mapping and similarity calculation for data from different media types. Nonetheless, this may cause information loss during the mapping process due to overlooking the specifics of each individual modality. In this paper, we propose a simple yet effective cross-modal hashing approach, termed collective reconstructive embeddings (CRE), which can simultaneously solve the heterogeneity and integration complexity of multi-modal data. To address the heterogeneity challenge, we propose to process heterogeneous types of data using different modality-specific models. Specifically, we model textual data with cosine similarity-based reconstructive embedding to alleviate the data sparsity to the greatest extent, while for image data, we utilize the Euclidean distance to characterize the relationships of the projected hash codes. Meanwhile, we unify the projections of text and image to the Hamming space into a common reconstructive embedding through rigid mathematical reformulation, which not only reduces the optimization complexity significantly but also facilitates the inter-modal similarity preservation among different modalities. We further incorporate the code balance and uncorrelation criteria into the problem and devise an efficient iterative algorithm for optimization. Comprehensive experiments on four widely used multimodal benchmarks show that the proposed CRE can achieve a superior performance compared with the state of the art on several challenging cross-modal tasks.

The effect of nanosilica on the sulfate attack resistivity of cement mortar was investigated through study on the mechanical property evolution and the length change of the cement mortar under 5 wt.% sodium sulfate for 6 months. Meanwhile, the effects were compared with those of fly ash-replacement mortar. Results showed that by taking the advantages of nanosilica and fly ash in improving the property of cement mortar at early and later ages, the sulfate attack resistance of cement mortar can be enhanced in mechanical property increase and expansion reduction. Further, it implies that a combination of both pozzolans could enhance the sulfate attack resistivity of cement-based materials.

We propose a framework for statistical modeling of the 3D geometry and topology of botanical trees. We treat botanical trees as points in a tree-shape space equipped with a proper metric that captures the geometric and the topological differences between trees. Geodesics in the tree-shape space correspond to the optimal sequence of deformations, i.e. bending, stretching, and topological changes, which align one tree onto another. In this way, the 3D tree modeling and synthesis problem becomes a problem of exploring the tree-shape space either in a controlled fashion, using statistical regression, or randomly by sampling from probability distributions fitted to populations in the tree-shape space. We show how to use this framework for (1) computing statistical summaries, e.g. the mean and modes of variations, of a population of botanical trees, (2) synthesizing random instances of botanical trees from probability distributions fitted to a population of botanical trees, and (3) modeling, interactively, 3D botanical trees using a simple sketching interface. The approach is fast and only requires as input 3D botanical tree models with a known upright orientation.

We propose an algorithm for generating novel 3D tree model variations from existing ones via geometric and structural blending. Our approach is to treat botanical trees as elements of a tree-shape space equipped with a proper metric that quantifies geometric and structural deformations. Geodesics, or shortest paths under the metric, between two points in the tree-shape space correspond to optimal deformations that align one tree onto another, including the possibility of expanding, adding, or removing branches and parts. Central to our approach is a mechanism for computing correspondences between trees that have different structures and a different number of branches. The ability to compute geodesics and their lengths enables us to compute continuous blending between botanical trees, which, in turn, facilitates statistical analysis, such as the computation of averages of tree structures. We show a variety of 3D tree models generated with our approach from 3D trees exhibiting complex geometric and structural differences. We also demonstrate the application of the framework in reflection symmetry analysis and symmetrization of botanical trees.

We propose a framework for statistical modeling of the 3D geometry and topology of botanical trees. We treat botanical trees as points in a tree-shape space equipped with a proper metric that captures the geometric and the topological differences between trees. Geodesics in the tree-shape space correspond to the optimal sequence of deformations, i.e. bending, stretching, and topological changes, which align one tree onto another. In this way, the 3D tree modeling and synthesis problem becomes a problem of exploring the tree-shape space either in a controlled fashion, using statistical regression, or randomly by sampling from probability distributions fitted to populations in the tree-shape space. We show how to use this framework for (1) computing statistical summaries, e.g. the mean and modes of variations, of a population of botanical trees, (2) synthesizing random instances of botanical trees from probability distributions fitted to a population of botanical trees, and (3) modeling, interactively, 3D botanical trees using a simple sketching interface. The approach is fast and only requires as input 3D botanical tree models with a known upright orientation.

We propose an algorithm for generating novel 3D tree model variations from existing ones via geometric and structural blending. Our approach is to treat botanical trees as elements of a tree-shape space equipped with a proper metric that quantifies geometric and structural deformations. Geodesics, or shortest paths under the metric, between two points in the tree-shape space correspond to optimal deformations that align one tree onto another, including the possibility of expanding, adding, or removing branches and parts. Central to our approach is a mechanism for computing correspondences between trees that have different structures and a different number of branches. The ability to compute geodesics and their lengths enables us to compute continuous blending between botanical trees, which, in turn, facilitates statistical analysis, such as the computation of averages of tree structures. We show a variety of 3D tree models generated with our approach from 3D trees exhibiting complex geometric and structural differences. We also demonstrate the application of the framework in reflection symmetry analysis and symmetrization of botanical trees.

The advantages and disadvantages of giving a valid name to a sequence of DNA detected from environmental specimens is presently a hot debate amongst the mycological community. The idea of using intracellular DNA (“mgDNA”) from environmental samples as holotypes seems at face value, to be a good idea, considering the expansion of knowledge among these ‘dark taxa’ or ‘dark matter fungi’ that it could provide (i.e. sequence based taxa without physical specimens and formal nomenclature). However, the limitations of using mgDNA as holotypes needs careful thought, i.e. can we use a short mgDNA fragment, which may contain a small amount of genetic information, to allow discrimination between species? What is the point and are the potential problems of giving valid scientific names to mgDNA? Numerous mycologists and taxonomists, who have many years of experience working on the taxonomy and phylogeny of different groups of fungi, are concerned about the consequences of providing valid names to mgDNA. There has been much debate, through several publications on the considerable problems of using mgDNA as holotypes. The proponents have tried to debate the virtues of using mgDNA as holotypes. Those against have shown that identification to species using mgDNA does not work in many fungal groups, while those for have shown cases where species can be identified with mgDNA. Different disciplines have different reasons and opinions for using mgDNA as holotypes, however even groups of the same disciplines have dissimilar ideas. In this paper we explore the use of mgDNA as holotypes. We provide evidences and opinions as to the use of mgDNA as holotypes from our own experiences. In no way do we attempt to degrade the study of DNA from environmental samples and the expansion of knowledge in to the dark taxa, but relate the issues to fungal taxonomy. In fact we show the value of using sequence data from these approaches, in dealing with the discovery of already named taxa, taxa numbers a

© 1992-2012 IEEE. Large-scale search methods are increasingly critical for many content-based visual analysis applications, among which hashing-based approximate nearest neighbor search techniques have attracted broad interests due to their high efficiency in storage and retrieval. However, existing hashing works are commonly designed for measuring data similarity by the Euclidean distances. In this paper, we focus on the problem of learning compact binary codes using the cosine similarity. Specifically, we proposed novel angular reconstructive embeddings (ARE) method, which aims at learning binary codes by minimizing the reconstruction error between the cosine similarities computed by original features and the resulting binary embeddings. Furthermore, we devise two efficient algorithms for optimizing our ARE in continuous and discrete manners, respectively. We extensively evaluate the proposed ARE on several large-scale image benchmarks. The results demonstrate that ARE outperforms several state-of-the-art methods.

© 1992-2012 IEEE. Large-scale search methods are increasingly critical for many content-based visual analysis applications, among which hashing-based approximate nearest neighbor search techniques have attracted broad interests due to their high efficiency in storage and retrieval. However, existing hashing works are commonly designed for measuring data similarity by the Euclidean distances. In this paper, we focus on the problem of learning compact binary codes using the cosine similarity. Specifically, we proposed novel angular reconstructive embeddings (ARE) method, which aims at learning binary codes by minimizing the reconstruction error between the cosine similarities computed by original features and the resulting binary embeddings. Furthermore, we devise two efficient algorithms for optimizing our ARE in continuous and discrete manners, respectively. We extensively evaluate the proposed ARE on several large-scale image benchmarks. The results demonstrate that ARE outperforms several state-of-the-art methods.

© 2018 Elsevier Inc. Among various traditional art forms, brush stroke drawing is one of the widely used styles in modern computer graphic tools such as GIMP, Photoshop and Painter. In this paper, we develop an AI-aided art authoring (A4) system of non-photorealistic rendering that allows users to automatically generate brush stroke paintings in a specific artist's style. Within the reinforcement learning framework of brush stroke generation proposed by Xie et al. (2012), the first contribution in this paper is the application of regularized policy gradient method, which is more suitable for the stroke generation task; the other contribution is to learn artists’ drawing styles from video-captured stroke data by inverse reinforcement learning. Through experiments, we demonstrate that our system can successfully learn artists’ styles and render pictures with consistent and smooth brush strokes.

© 2018 Elsevier Inc. Among various traditional art forms, brush stroke drawing is one of the widely used styles in modern computer graphic tools such as GIMP, Photoshop and Painter. In this paper, we develop an AI-aided art authoring (A4) system of non-photorealistic rendering that allows users to automatically generate brush stroke paintings in a specific artist's style. Within the reinforcement learning framework of brush stroke generation proposed by Xie et al. (2012), the first contribution in this paper is the application of regularized policy gradient method, which is more suitable for the stroke generation task; the other contribution is to learn artists’ drawing styles from video-captured stroke data by inverse reinforcement learning. Through experiments, we demonstrate that our system can successfully learn artists’ styles and render pictures with consistent and smooth brush strokes.

© 2017 IEEE. Nowadays, voice acting plays more advanced in the video games, especially for the role-playing games, anime-based games and serious games. In order to enhance the communication, synchronizing the lip and mouth movements naturally is an important part of convincing 3D character performance [XFMS13]. In this paper, we propose a lightweight LipSync generation algorithm. According to the heuristic knowledge on the mouth movement in game, extracting the value of voice frequency domain is essential for LipSync in game. Therefore, we analytically convert the problem into Discrete Cosine Transform (DCT) that focuses on extracting the voice frequency domain value by absolute value computing operation so as to avoid redundant computation on phases and modulus of operation in Fourier Transform (FT) voice model. Our experimental results demonstrate that our DCT based method enables to achieve good performance for game making.

© 2017 Association for Computing Machinery. Subspace representations have been widely applied for videos in many tasks. In particular, the subspace-based query-by-image video retrieval (QBIVR), facing high challenges on similarity-preserving measurements and efficient retrieval schemes, urgently needs considerable research attention. In this paper, we propose a novel subspace-based QBIVR framework to enable efficient video search. We first define a new geometry-preserving distance metric to measure the image-to-video distance, which transforms the QBIVR task to be the Maximum Inner Product Search (MIPS) problem. The merit of this distance metric lies in that it helps to preserve the genuine geometric relationship between query images and database videos to the greatest extent. To boost the efficiency of solving the MIPS problem, we introduce two asymmetric hashing schemes which can bridge the domain gap of images and videos properly. The first approach, termed Inner-product Binary Coding (IBC), achieves high-quality binary codes by learning the binary codes and coding functions simultaneously without continuous relaxations. The other one, Bilinear Binary Coding (BBC) approach, employs compact bilinear projections instead of a single large projection matrix to further improve the retrieval efficiency. Extensive experiments on four real-world video datasets verify the effectiveness of our proposed approaches, as compared to the state-of-the-art methods.

© 2017, Springer International Publishing AG. Computer-aided drawing system assists users to convert the input real photos into painterly style images. Nowadays, it is widely developed as cloud brush engine service in many creative software tools and applications of artistic rendering such as Prisma [1], Photoshop [2], and Meitu [3], because the machine learning server has more powerful than the stand-alone version. In this paper, we propose a web collaborative Stroke-based Learning and Rendering (WebSBLR) system. Different from the existing methods that are mainly focused on the artistic filters, we concentrate on the stroke realistic rendering engine for browser on client using WebGL and HTML5. Moreover, we implement the learning-based stroke drawing path generation module on the server. By this way, we enable to achieve the computer-supported cooperative work (CSCW), especially for multi-screen synchronous interaction. The experiments demonstrated our method are efficient to web-based multi-screen painting simulation.

© 2017, Springer International Publishing AG. With the explosive growth of user-generated contents (e.g., texts, images and videos) on social networks, it is of great significance to analyze and extract people’s interests from the massive social media data, thus providing more accurate personalized recommendations and services. In this paper, we propose a novel multimodal deep learning algorithm for user profiling, dubbed multi-modal User Attribute Model (mmUAM), which explores the intrinsic semantic correlations across different modalities. Our proposed model is based on Poisson Gamma Belief Network (PGBN), which is a deep learning topic model for count data in documents. By improving PGBN, we succeed in addressing the problem of learning a shared representation between texts and images in order to obtain textual and visual attributes for users. To evaluate the effectiveness of our proposed method, we collect a real dataset from Sina Weibo. Experimental results demonstrate that the proposed algorithm achieves encouraging performance compared with several state-of-the-art methods.

© 2017 Copyright held by the owner/author(s). Cosmetic medical visualization has become an important application in computer graphics, especially for facial appearance visualization[Chandawarkar et al. 2013]. Recent approaches have reached very realistic results by blend shape[Ma et al. 2012], which is the most practical tool to make the facial appearance and expression animation in application domains on the entertainment industry (VFXs and games). In many role-playing games (RPGs), players enable to edit the character's facial appearance. However, it is unrealistic since arbitrary discontinuities and position relationship violations (a selected nose might be at a higher position that the bottom of the eyes selected from a different character) caused by players' manual operation. Moreover, the validity on changing facial organs has not be considered well yet.

© 2017 Copyright held by the owner/author(s). Predicting the popularity of Point of Interest (POI) has be-come increasingly crucial for location-based services, such as POI recommendation. Most of the existing methods can seldom achieve satisfactory performance due to the scarci-Ty of POI's information, which tendentiously confines the recommendation to popular scenic spots, and ignores the unpopular attractions with potentially precious values. In this paper, we propose a novel approach, termed Hierarchical Multi-Clue Fusion (HMCF), for predicting the popularity of POIs. Specifically, we devise an effective hierarchy to comprehensively describe POI by integrating various types of media information (e.g., image and text) from multiple social sources. For each individual POI, we simultaneously inject se-mantic knowledge as well as multi-clue representative power. We collect a multi-source POI dataset from four widely-used tourism platforms. Extensive experimental results show that the proposed method can significantly improve the perfor-mance of predicting the attractions' popularity as compared to several baselines.

© 2017 Copyright held by the owner/author(s). Cosmetic medical visualization has become an important application in computer graphics, especially for facial appearance visualization[Chandawarkar et al. 2013]. Recent approaches have reached very realistic results by blend shape[Ma et al. 2012], which is the most practical tool to make the facial appearance and expression animation in application domains on the entertainment industry (VFXs and games). In many role-playing games (RPGs), players enable to edit the character's facial appearance. However, it is unrealistic since arbitrary discontinuities and position relationship violations (a selected nose might be at a higher position that the bottom of the eyes selected from a different character) caused by players' manual operation. Moreover, the validity on changing facial organs has not be considered well yet.

© 2017, Springer International Publishing AG. Computer-aided drawing system assists users to convert the input real photos into painterly style images. Nowadays, it is widely developed as cloud brush engine service in many creative software tools and applications of artistic rendering such as Prisma [1], Photoshop [2], and Meitu [3], because the machine learning server has more powerful than the stand-alone version. In this paper, we propose a web collaborative Stroke-based Learning and Rendering (WebSBLR) system. Different from the existing methods that are mainly focused on the artistic filters, we concentrate on the stroke realistic rendering engine for browser on client using WebGL and HTML5. Moreover, we implement the learning-based stroke drawing path generation module on the server. By this way, we enable to achieve the computer-supported cooperative work (CSCW), especially for multi-screen synchronous interaction. The experiments demonstrated our method are efficient to web-based multi-screen painting simulation.

© 2017, Springer International Publishing AG. With the explosive growth of user-generated contents (e.g., texts, images and videos) on social networks, it is of great significance to analyze and extract people’s interests from the massive social media data, thus providing more accurate personalized recommendations and services. In this paper, we propose a novel multimodal deep learning algorithm for user profiling, dubbed multi-modal User Attribute Model (mmUAM), which explores the intrinsic semantic correlations across different modalities. Our proposed model is based on Poisson Gamma Belief Network (PGBN), which is a deep learning topic model for count data in documents. By improving PGBN, we succeed in addressing the problem of learning a shared representation between texts and images in order to obtain textual and visual attributes for users. To evaluate the effectiveness of our proposed method, we collect a real dataset from Sina Weibo. Experimental results demonstrate that the proposed algorithm achieves encouraging performance compared with several state-of-the-art methods.

© 2017 Copyright held by the owner/author(s). Predicting the popularity of Point of Interest (POI) has be-come increasingly crucial for location-based services, such as POI recommendation. Most of the existing methods can seldom achieve satisfactory performance due to the scarci-Ty of POI's information, which tendentiously confines the recommendation to popular scenic spots, and ignores the unpopular attractions with potentially precious values. In this paper, we propose a novel approach, termed Hierarchical Multi-Clue Fusion (HMCF), for predicting the popularity of POIs. Specifically, we devise an effective hierarchy to comprehensively describe POI by integrating various types of media information (e.g., image and text) from multiple social sources. For each individual POI, we simultaneously inject se-mantic knowledge as well as multi-clue representative power. We collect a multi-source POI dataset from four widely-used tourism platforms. Extensive experimental results show that the proposed method can significantly improve the perfor-mance of predicting the attractions' popularity as compared to several baselines.

© 2017 IEEE. Nowadays, voice acting plays more advanced in the video games, especially for the role-playing games, anime-based games and serious games. In order to enhance the communication, synchronizing the lip and mouth movements naturally is an important part of convincing 3D character performance [XFMS13]. In this paper, we propose a lightweight LipSync generation algorithm. According to the heuristic knowledge on the mouth movement in game, extracting the value of voice frequency domain is essential for LipSync in game. Therefore, we analytically convert the problem into Discrete Cosine Transform (DCT) that focuses on extracting the voice frequency domain value by absolute value computing operation so as to avoid redundant computation on phases and modulus of operation in Fourier Transform (FT) voice model. Our experimental results demonstrate that our DCT based method enables to achieve good performance for game making.

© 2017 Association for Computing Machinery. Subspace representations have been widely applied for videos in many tasks. In particular, the subspace-based query-by-image video retrieval (QBIVR), facing high challenges on similarity-preserving measurements and efficient retrieval schemes, urgently needs considerable research attention. In this paper, we propose a novel subspace-based QBIVR framework to enable efficient video search. We first define a new geometry-preserving distance metric to measure the image-to-video distance, which transforms the QBIVR task to be the Maximum Inner Product Search (MIPS) problem. The merit of this distance metric lies in that it helps to preserve the genuine geometric relationship between query images and database videos to the greatest extent. To boost the efficiency of solving the MIPS problem, we introduce two asymmetric hashing schemes which can bridge the domain gap of images and videos properly. The first approach, termed Inner-product Binary Coding (IBC), achieves high-quality binary codes by learning the binary codes and coding functions simultaneously without continuous relaxations. The other one, Bilinear Binary Coding (BBC) approach, employs compact bilinear projections instead of a single large projection matrix to further improve the retrieval efficiency. Extensive experiments on four real-world video datasets verify the effectiveness of our proposed approaches, as compared to the state-of-the-art methods.

There is growing interest in the notion that perioperative factors may influence the long-term outcome of cancer surgery. Among these factors, pharmacologic agents including anesthetics and analgesics have been scrutinized for their potential effect on tumors. Opioids are particularly interesting in this context, as conflicting literature reports both pro- and anti-tumor effects for this class of drugs in every aspect of tumor growth and metastasis that has been examined. Cancer and non-cancer cells important to the tumor biology may be affected, as demonstrated using a variety of in vitro and preclinical models. More evidence is required to optimize the perioperative pain management of cancer surgery patients.

There is growing interest in the notion that perioperative factors may influence the long-term outcome of cancer surgery. Among these factors, pharmacologic agents including anesthetics and analgesics have been scrutinized for their potential effect on tumors. Opioids are particularly interesting in this context, as conflicting literature reports both pro- and anti-tumor effects for this class of drugs in every aspect of tumor growth and metastasis that has been examined. Cancer and non-cancer cells important to the tumor biology may be affected, as demonstrated using a variety of in vitro and preclinical models. More evidence is required to optimize the perioperative pain management of cancer surgery patients.

A micro-digital sun sensor (?DSS) is a sun detector which senses a satellite’s instant attitude angle with respect to the sun. The core of this sensor is a system-on-chip imaging chip which is referred to as APS+. The APS+ integrates a CMOS active pixel sensor (APS) array of 368×368??pixels , a 12-bit analog-to-digital converter, and digital signal processing circuits. The ?DSS is designed particularly for microsatellite applications, thus low power consumption is the major design consideration. The APS+ reduces power consumption mainly with profiling and windowing methods which are facilitated by the specific active-pixel design. A prototype of the APS+ which is designed in a standard 0.18-?m CMOS process is presented. The APS+ consumes 21 mW at 10 fps, which is 10 times less than the state of the art. In order to improve noise performance, a reset noise reduction method, quadruple sampling (QS), is implemented. QS reduces the effect of the reset noise compared to the conventional delta double sampling method, even in a 3-transistor active pixel structure. The APS+ obtains an accuracy of 0.01 deg with the QS method., Microelectronics, Electrical Engineering, Mathematics and Computer Science

A micro-digital sun sensor (?DSS) is a sun detector which senses a satellite’s instant attitude angle with respect to the sun. The core of this sensor is a system-on-chip imaging chip which is referred to as APS+. The APS+ integrates a CMOS active pixel sensor (APS) array of 368×368??pixels , a 12-bit analog-to-digital converter, and digital signal processing circuits. The ?DSS is designed particularly for microsatellite applications, thus low power consumption is the major design consideration. The APS+ reduces power consumption mainly with profiling and windowing methods which are facilitated by the specific active-pixel design. A prototype of the APS+ which is designed in a standard 0.18-?m CMOS process is presented. The APS+ consumes 21 mW at 10 fps, which is 10 times less than the state of the art. In order to improve noise performance, a reset noise reduction method, quadruple sampling (QS), is implemented. QS reduces the effect of the reset noise compared to the conventional delta double sampling method, even in a 3-transistor active pixel structure. The APS+ obtains an accuracy of 0.01 deg with the QS method., Microelectronics, Electrical Engineering, Mathematics and Computer Science

OBJECTIVE: To evaluate the application of health assessment instruments in Chinese medicine. METHODS: According to a pre-defined search strategy, a comprehensive literature search for all articles published in China National Knowledge Infrastructure databases was conducted. The resulting articles that met the defined inclusion and exclusion criteria were used for analysis. RESULTS: A total of 97 instruments for health outcome assessment in Chinese medicine have been used in fundamental and theoretical research, and 14 of these were also used in 29 clinical trials that were randomized controlled trials, or descriptive or cross-sectional studies. In 2 152 Chinese medicine-based studies that used instruments in their methodology, more than 150 questionnaires were identified. Among the identified questionnaires, 51 were used in more than 10 articles (0.5%). Most of these instruments were developed in Western countries and few studies (4%) used the instrument as the primary evidence for their conclusions. CONCLUSION: Usage of instruments for health outcome assessment in Chinese medicine is increasing rapidly; however, current limitations include selection rationale, result interpretation and standardization, which must be addressed accordingly.

OBJECTIVE: To evaluate the development of health outcomes assessment instruments in Chinese medicine. METHODS: A comprehensive literature search for all published articles in China National Knowledge Infrastructure Database, Chongqing VIP Database and WANFANG Data was conducted. The studies that met the inclusion and exclusion criteria were used to extract information according to a predesigned assessment instrument. RESULTS: A total of 97 instruments for health outcome assessment in Chinese medicine were identified. Of these questionnaires, 7 were generic, 12 were condition-specific and 78 were disease-specific. All instruments were suitable for adults, children, and both men and women. These instruments aimed to evaluate the health-related quality of life, signs and symptoms as well as patient satisfaction and doctor-reported outcome. However, the descriptions were poorly constructed for some of the most basic parameters, such as the domains and items, administrative mode, response options, memory recall periods, burden evaluation, format, copyright, content validity, and other properties. CONCLUSION: The instrument development for health outcomes assessment in Chinese medicine is increasing rapidly; however, there are many limitations in current methodologies and standards, and further studies are needed. © 2013 Feng-bin Liu et al.

OBJECTIVE: To evaluate the application of health assessment instruments in Chinese medicine. METHODS: According to a pre-defined search strategy, a comprehensive literature search for all articles published in China National Knowledge Infrastructure databases was conducted. The resulting articles that met the defined inclusion and exclusion criteria were used for analysis. RESULTS: A total of 97 instruments for health outcome assessment in Chinese medicine have been used in fundamental and theoretical research, and 14 of these were also used in 29 clinical trials that were randomized controlled trials, or descriptive or cross-sectional studies. In 2 152 Chinese medicine-based studies that used instruments in their methodology, more than 150 questionnaires were identified. Among the identified questionnaires, 51 were used in more than 10 articles (0.5%). Most of these instruments were developed in Western countries and few studies (4%) used the instrument as the primary evidence for their conclusions. CONCLUSION: Usage of instruments for health outcome assessment in Chinese medicine is increasing rapidly; however, current limitations include selection rationale, result interpretation and standardization, which must be addressed accordingly.

OBJECTIVE: To evaluate the development of health outcomes assessment instruments in Chinese medicine. METHODS: A comprehensive literature search for all published articles in China National Knowledge Infrastructure Database, Chongqing VIP Database and WANFANG Data was conducted. The studies that met the inclusion and exclusion criteria were used to extract information according to a predesigned assessment instrument. RESULTS: A total of 97 instruments for health outcome assessment in Chinese medicine were identified. Of these questionnaires, 7 were generic, 12 were condition-specific and 78 were disease-specific. All instruments were suitable for adults, children, and both men and women. These instruments aimed to evaluate the health-related quality of life, signs and symptoms as well as patient satisfaction and doctor-reported outcome. However, the descriptions were poorly constructed for some of the most basic parameters, such as the domains and items, administrative mode, response options, memory recall periods, burden evaluation, format, copyright, content validity, and other properties. CONCLUSION: The instrument development for health outcomes assessment in Chinese medicine is increasing rapidly; however, there are many limitations in current methodologies and standards, and further studies are needed. © 2013 Feng-bin Liu et al.

A digital sun sensor is superior to an analog sun sensor in aspects of resolution, albedo immunity, and integration. The proposed Micro-Digital Sun Sensor (µDSS) is an autonomous digital sun sensor which is implemented by means of a CMOS image sensor, which is named APS+. The µDSS is designed specifically for micro-satellite application which addresses for low power and high accuracy. The APS+ significantly reduces the power consumption by a profiling method and obtains low noise by a quadruple sampling method. Measurement results show that the µDSS achieves an accuracy of 0.01º with 21mW@10fps, which is 10 times less than the state of the art., Microelectronics & Computer Engineering, Electrical Engineering, Mathematics and Computer Science

A digital sun sensor is superior to an analog sun sensor in aspects of resolution, albedo immunity, and integration. The proposed Micro-Digital Sun Sensor (µDSS) is an autonomous digital sun sensor which is implemented by means of a CMOS image sensor, which is named APS+. The µDSS is designed specifically for micro-satellite application which addresses for low power and high accuracy. The APS+ significantly reduces the power consumption by a profiling method and obtains low noise by a quadruple sampling method. Measurement results show that the µDSS achieves an accuracy of 0.01º with 21mW@10fps, which is 10 times less than the state of the art., Microelectronics & Computer Engineering, Electrical Engineering, Mathematics and Computer Science

Micro-Digital Sun Sensor (?DSS) is a sun detector which senses the respective angle between a satellite and the sun. It is composed of a solar cell power supply, a RF communication block and a CMOS Image Sensor (CIS) chip, which is called APS+. The paper describes the implementation of a prototype of the ?DSS APS+ processed in a standard 0.18?m CMOS process. The ?DSS is applied for micro or nano satellites. Power consumption is a very rigid specification in this kind of application, thus the APS+ is optimized for low power consumption. This character is realized by a specific pixel design which implements profiling and windowing during the detection process. The profiling is completely fast and power efficiently by a “Winner Take ALL (WTA)” principle. The measurement results shows that the APS+ achieves a reduction of power consumption by more than a factor 10 compared to state of-the-art. Besides the low power consumption, the APS+ also proposes a quadruple sampling method which improves thermal noise with 3-T Active Pixel image Sensor (APS) structure., Microelectronics, Electrical Engineering, Mathematics and Computer Science

Micro-Digital Sun Sensor (?DSS) is a sun detector which senses the respective angle between a satellite and the sun. It is composed of a solar cell power supply, a RF communication block and a CMOS Image Sensor (CIS) chip, which is called APS+. The paper describes the implementation of a prototype of the ?DSS APS+ processed in a standard 0.18?m CMOS process. The ?DSS is applied for micro or nano satellites. Power consumption is a very rigid specification in this kind of application, thus the APS+ is optimized for low power consumption. This character is realized by a specific pixel design which implements profiling and windowing during the detection process. The profiling is completely fast and power efficiently by a “Winner Take ALL (WTA)” principle. The measurement results shows that the APS+ achieves a reduction of power consumption by more than a factor 10 compared to state of-the-art. Besides the low power consumption, the APS+ also proposes a quadruple sampling method which improves thermal noise with 3-T Active Pixel image Sensor (APS) structure., Microelectronics, Electrical Engineering, Mathematics and Computer Science