IM(S)2: Interactive movie summarization system

The need of summarization methods and systems has become more and more crucial as the audio-visual material continues its critical growth. This paper presents a novel vision and a novel system for movies summarization. A video summary is an audio-visual document displaying the essential parts of an original document. However, the definition of the term “essential” is user-dependent. The advantage of this work, unlike the others, is the involvement of users in the summarization process. By means of IM(S)², people generate on the fly customized video summaries responding to their preferences. IM(S)² is made up of an offline part and an online part. In the offline, we segment the movies into shots and we compute features describing them. In the online part users inform about their preferences by selecting interesting shots. After that, the system will analyze the selected shots to bring out the user’s preferences. Finally the system will generate a summary from the whole movie which will provide more focus on the user’s preferences. To show the efficiency of IM(S)², it was tested on the database of the European project MUSCLE made up of five movies. We invited 10 users to evaluate the usability of our system by generating for every movie of the database a semi-supervised summary and to judge at the end its quality. Obtained results are encouraging and show the merits of our approach.     

一种基于用户动态兴趣和社交网络的微博推荐方法

针对为微博用户推荐符合其兴趣取向的个性化微博信息的问题,结合LDA主题模型,提出了一种基于用户动态兴趣和社交网络（DISN）的微博推荐方法.DISN方法首先引入时间函数,推断出用户的兴趣向量,通过对新发布的微博数据内容进行聚类分组,以用户兴趣向量筛选与用户最匹配的分组,随后以网格索引的形式对选定的分组中微博进行查询,计算微博发布者被目标用户关注的可能性并进行排序,最终形成推荐列表.实验验证了DISN方法较之传统方法更具有效性和高效性.     

基于用户主题模型的微博用户兴趣挖掘(英文)

Microblogs have become an important platform for people to publish,transform information and acquire knowledge.This paper focuses on the problem of discovering user interest in microblogs.In this paper,we propose a topic mining model based on Latent Dirichlet Allocation(LDA) named user-topic model.For each user,the interests are divided into two parts by different ways to generate the microblogs:original interest and retweet interest.We represent a Gibbs sampling implementation for inference the parameters of our model,and discover not only user's original interest,but also retweet interest.Then we combine original interest and retweet interest to compute interest words for users.Experiments on a dataset of Sina microblogs demonstrate that our model is able to discover user interest effectively and outperforms existing topic models in this task.And we find that original interest and retweet interest are similar and the topics of interest contain user labels.The interest words discovered by our model reflect user labels,but range is much broader.     

External Expansion Risk Management: Enhancing Microblogging Filtering Using Implicit Query

Microblogging filtering can help users filter out irrelevant content, and extract timely content effectively from microblogs. However, as a typical short text, microblogging filtering suffers from the insufficient samples problem that makes the probabilistic-like models unreliable. According to the current research, an explicit brief query has been thought to be only an abstract of the user’s information needs, and it’s hard to infer what is the users’ actual searching intents. Instead, we submit the relevant external documents as a user’s implicit prior knowledge and then build a corresponding filtering framework. To against the risk of external documents expansion, we suppose the external document can be viewed as a complete statement of an explicit query, and encode the filtering preferences with the diverge degree between the external document and the the original explicit query. Thus the optimal filtering action is the one that allows one to trade off diverge degree against generalization performance. With respect to the established baselines, our algorithm yields compelling results for providing a meaningful tweets retrieval. This work helps further understand the innate risk characteristics of external expansion for the design of Microblogging filtering systems.     

Using MPEG-7 and MPEG-21 for personalizing video

As multimedia content has proliferated over the past several years, users have begun to expect that content be easily accessed according to their own preferences. One of the most effective ways to do this is through using the MPEG-7 and MPEG-21 standards, which can help address the issues associated with designing a video personalization and summarization system in heterogeneous usage environments. This three-tier architecture provides a standards-compliant infrastructure that, in conjunction with our tools, can help select, adapt, and deliver personalized video summaries to users. In extending our summarization research, we plan to explore semantic similarities across multiple simultaneous news media sources and to abstract summaries for different viewpoints. Doing so will allow us to track a semantic topic as it evolves into the future. As a result, we should be able to summarize news repositories into a smaller collection of topic threads.     

Cross-Layer Optimization for Video Summary Transmission over Wireless Networks

Video summarization has gained increased popularity in the emerging multimedia communication applications, however, very limited work has been conducted to address the transmission problem of video summary frames. In this paper, we propose a cross-layer optimization framework for delivering video summaries over wireless networks. Within a rate-distortion theoretical framework, the source coding, allowable retransmission, and adaptive modulation and coding have been jointly optimized, which reflects the joint selection of parameters at physical, data link and application layers. The goal is to achieve the best video quality and content coverage of the received summary frames and to meet the delay constraint. The problem is solved using Lagirangian relaxation and dynamic programming. Experimental results indicate the effectiveness and efficiency of the proposed optimization framework, especially when the delay budget imposed by the upper layer applications is small, where more than 10% distortion gain can be achieved.     

一种基于Opinosis图和马尔科夫随机游走模型的多文本情感摘要框架

针对在线文本情感摘要生成问题,本文提出了一种基于Opinosis图和马尔科夫随机游走模型的情感摘要框架.首先,该框架将原始文本转化为Opinosis图,并利用其挖掘出文本中的特征词,这些特征词可以用来对原始文本的句子进行分类;其次本文在基于聚类的条件马尔科夫随机游走模型的基础上增加了情感层,改进后的模型可以判断同一聚类中各句子的情感倾向是否具有代表性并结合情感和聚类信息对句子进行排序.实验结果表明,本文提出的方法与基准算法相比在ROUGE（Recall-Oriented Understudy for Gisting Evaluation）值上具有明显提高.     

Memorable and rich video summarization

Video summarization can facilitate rapid browsing and efficient video indexing in many applications. A good summary should maintain the semantic interestingness and diversity of the original video. While many previous methods extracted key frames based on low-level features, this study proposes Memorability-Entropy-based video summarization. The proposed method focuses on creating semantically interesting summaries based on image memorability. Further, image entropy is introduced to maintain the diversity of the summary. In the proposed framework, perceptual hashing-based mutual information (MI) is used for shot segmentation. Then, we use a large annotated image memorability dataset to fine-tune Hybrid-AlexNet. We predict the memorability score by using the fine-tuned deep network and calculate the entropy value of the images. The frame with the maximum memorability score and entropy value in each shot is selected to constitute the video summary. Finally, our method is evaluated on a benchmark dataset, which comes with five human-created summaries. When evaluating our method, we find it generates high-quality results, comparable to human-created summaries and conventional methods.     

基于限制性玻尔兹曼机的微博主题分类

智慧城市依赖于对大数据的充分利用。近年来,随着移动互联网的发展,在线微博平台,比如新浪微博、Twitter等,已经成为了大数据的主要来源之一。微博平台上产生的海量短文文本信息使用户很难找到自己感兴趣主题的相关信息。本文提出了一种基于限制性玻尔兹曼机的微博短文本的主题分类方法。通过对短文本进行主题建模,挖掘出潜在主题信息,根据短文本的潜在主题信息可实现对短文本的主题分类。     

Two-stage sentence selection approach for multi-document summarization

Compared with the traditional method of adding sentences to get summary in multi-document summarization,a two-stage sentence selection approach based on deleting sentences in a candidate sentence set to generate summary is proposed,which has two stages,the acquisition of a candidate sentence set and the optimum selection of sentence.At the first stage,the candidate sentence set is obtained by redundancy-based sentence selection approach.At the second stage,optimum selection of sentences is proposed to delete sentences in the candidate sentence set according to its contribution to the whole set until getting the appointed summary length.With a test corpus,the ROUGE value of summaries gotten by the proposed approach proves its validity,compared with the traditional method of sentence selection.The influence of the token chosen in the two-stage sentence selection approach on the quality of the generated summaries is analyzed.     

Automatic soccer video analysis and summarization

We propose a fully automatic and computationally efficient framework for analysis and summarization of soccer videos using cinematic and object-based features. The proposed framework includes some novel low-level processing algorithms, such as dominant color region detection, robust shot boundary detection, and shot classification, as well as some higher-level algorithms for goal detection, referee detection, and penalty-box detection. The system can output three types of summaries: i) all slow-motion segments in a game; ii) all goals in a game; iii) slow-motion segments classified according to object-based features. The first two types of summaries are based on cinematic features only for speedy processing, while the summaries of the last type contain higher-level semantics. The proposed framework is efficient, effective, and robust. It is efficient in the sense that there is no need to compute object-based features when cinematic features are sufficient for the detection of certain events, e.g., goals in soccer. It is effective in the sense that the framework can also employ object-based features when needed to increase accuracy (at the expense of more computation). The efficiency, effectiveness, and robustness of the proposed framework are demonstrated over a large data set, consisting of more than 13 hours of soccer video, captured in different countries and under different conditions.     

Adaptive key frame extraction for video summarization using an aggregation mechanism

Video summarization is a method to reduce redundancy and generate succinct representation of the video data. One of the mechanisms to generate video summaries is to extract key frames which represent the most important content of the video. In this paper, a new technique for key frame extraction is presented. The scheme uses an aggregation mechanism to combine the visual features extracted from the correlation of RGB color channels, color histogram, and moments of inertia to extract key frames from the video. An adaptive formula is then used to combine the results of the current iteration with those from the previous. The use of the adaptive formula generates a smooth output function and also reduces redundancy. The results are compared to some of the other techniques based on objective criteria. The experimental results show that the proposed technique generates summaries that are closer to the summaries created by humans.     

EcoPlan: energy-efficient downlink and uplink data transmission in mobile cloud computing

Mobile cloud computing (MC2) is emerging as a new computing paradigm that seeks to augment resource-constrained mobile devices for executing computing- and/or data-intensive mobile applications. Nonetheless, the energy-poverty nature of mobile devices has become a stumbling block that greatly impedes the practical application of MC2. Fortunately, for delay-tolerant mobile applications, energy conservation is achievable via two means: (1) dynamic selection of energy-efficient links (e.g., WiFi interface); and (2) deferring data transmission in bad connectivity. In this paper, we study the problem of energy-efficient downlink and uplink data transmission between mobile devices and clouds. In the presence of unpredictable data arrival, network availability and link quality, our objective is to minimize the time average energy consumption of a mobile device while ensuring the stability of both device-end and cloud-end queues. To achieve this goal, we propose an online control framework named EcoPlan under which mobile users can make flexible link selection and data transmission scheduling decisions to achieve arbitrary energy-delay tradeoffs. Real-world trace-driven simulations demonstrate the effectiveness of EcoPlan, along with its superior energy-efficiency over alternative WiFi-prioritized, minimum-delay and SALSA schemes.     

Generating visual story graphs with application to photo album summarization

Making sense of ever-growing amount of visual data available on the web is difficult, especially when considered in an unsupervised manner. As a step towards this goal, this study tackles a relatively less explored topic of generating structured summaries of large photo collections. Our framework relies on the notion of a story graph which captures the main narratives in the data and their relationships based on their visual, textual and spatio-temporal features. Its output is a directed graph with a set of possibly intersecting paths. Our proposed approach identifies coherent visual storylines and exploits sub-modularity to select a subset of these lines which covers the general narrative at most. Our experimental analysis reveals that extracted story graphs allow for obtaining better results when utilized as priors for photo album summarization. Moreover, our user studies show that our approach delivers better performance on next image prediction and coverage tasks than the state-of-the-art.     

A comprehensive survey and mathematical insights towards video summarization

Video Summarization is a technique to reduce the original raw video into a short video summary. Video summarization automates the task of acquiring key frames/segments from the video and combining them to generate a video summary. This paper provides a framework for summarization based on different criteria and also compares different literature work related to video summarization. The framework deals with formulating model for video summarization based on different criteria. Based on target audience/ viewership, number of videos, type of output intended, type of video summary and summarization factor; a model generating video summarization framework is proposed. The paper examines significant research works in the area of video summarization to present a comprehensive review against the framework. Different techniques, perspectives and modalities are considered to preserve the diversity of survey. This paper examines important mathematical formulations to provide meaningful insights for video summarization model creation.     

Dynamic modeling approach to analyze user churn in a large P2P downloading system

Characterizing the features of user churn is crucial to the sustainable development of peer-to-peer (P2P) systems where peers join and leave at any arbitrary time. This paper analyzes the user churn in a P2P downloading system named QQXuanfeng by using the fine-grained log analysis over 60 days. It shows that the online and offline duration is related to up (arrive) time and down (depart) time respectively. A continuous ON/OFF process, which exhibits the diurnal patterns of users, is simulated using the churn model. In particular, the dynamic departure rate is proposed to give insight into the distribution of online duration. Further more, considering the heterogeneity of users, we cluster users based on the similarity of redefined user availability. As an example of application of this model, a high availability overlay is constructed and evaluated based on the clustering.     

A Multi-Server Scheduling Framework for Resource Allocation in Wireless Multi-Carrier Networks

Multiuser resource allocation has recently been recognized as an effective methodology for enhancing the power and spectrum efficiency in OFDM (orthogonal frequency division multiplexing) systems. It is, however, not directly applicable to current packet-switched networks because most existing packet- scheduling schemes are based on a single-server model and do not serve multiple users at the same time. In this paper, we propose a cross-layer resource allocation algorithm based on a novel multi-server scheduling framework to achieve overall high system power efficiency in packet-switched OFDM networks. Our contribution is four fold: (i) we propose and analyze a MPGPS (multi-server packetized general processor sharing) service discipline that serves multiple users at the same time and facilitates multiuser resource allocation; (ii) we present a MPGPS-based joint MAC-PHY resource allocation scheme that incorporates packet scheduling, subcarrier allocation, and power allocation in an integrated framework; (iii) by investigating the fundamental tradeoff between multiuser-diversity and queueing performance, we present an A-MPGPS (adaptive MPGPS) service discipline that strikes balance between power efficiency and queueing performance; and (iv) we extend MPGPS to an O-MPGPS (opportunistic MPGPS) service discipline to further enhance the resource utilization efficiency.     

An integrated adaptive bandwidth-management framework for QoS-sensitive multimedia cellular networks

Bandwidth is an extremely valuable and scarce resource in a wireless network. Therefore, efficient bandwidth management is necessary in order to provide high-quality service to users in a multimedia wireless/mobile network. In this paper, we propose new online bandwidth-management algorithms for bandwidth reservation, call admission, bandwidth migration, and call-preemption strategies. These techniques are combined in an integrated framework that is able to balance the traffic load among cells accommodating heterogeneous multimedia services while ensuring efficient bandwidth utilization. In addition, our online framework to adaptively control bandwidth is a cell-oriented approach that has low complexity, which makes it practical for real cellular networks. Simulation results indicate the superior performance of our bandwidth-management framework to strike the appropriate performance balance between contradictory quality-of-service requirements.     

Temporal Topic-Based Multi-Dimensional Social Influence Evaluation in Online Social Networks

Analytical applications in online social networks can be generalized as the influence evaluation problem, which targets at finding most influential users. Nowadays social influence evaluation is still an open and challenging issue. Most influence evaluation models focus on the single dimensional evaluation factor but fail to research on the multi-dimensional factors. In this paper, we propose a novel influence evaluation model: the temporal topic influence (TTI) evaluation model, which is a time-aware, content-aware and structure-aware evaluation model. For the aim of multi-dimensional evaluation, we incorporate multi-dimensional measure factors into our model, including the time factor, the topological information and the topic distribution information, etc. We propose a novel concept of user gravitational ability which is inspired by Newton’s law of universal gravitation. It can integrate multi-dimensional factors in an appropriate way. Our experiments are conducted on the Sina Weibo data set. Through the experimental analysis, we prove TTI model can calculate users’ influences effectively and efficiently. The TTI model can distinguish the value of users’ influences. And the TTI model identifies the top-k influential users with higher quality. We also validate the effect of time and topic measure factors in the influence evaluation process.     

Preference-aware coding data broadcast in delay tolerant networks

Traditional data broadcasting schemes in delay tolerant networks assume that mobile users can only retrieve one data item in each time slot. In this paper, we propose a novel data broadcasting framework in the delay tolerant networks that exploits the concept of network coding to mix the delivered data items according to the user’s stored data items. Our approach enables a user to encode multiple data items dynamically in each time slot, and allows each user with a mobile device to retrieve a data item by using locally stored data items to decode the encoding data. Specifically, we design an algorithm called Preference-Aware Coding (PAC) to select the data items to be encoded in each time slot. The objective is to serve the maximal number of mobile users with the encoding data and minimize the access time required for data broadcasting in the delay tolerant networks. The algorithm avoids encoding unnecessary data in each time slot to reduce the access delay. We empirically implement the framework in the real delay tolerant networks, and simulation results show that the PAC algorithm can reduce the access time of the traditional scheme by 42 % on average.