20,398 results
Search Results
2. Architecture for Active Conceptual Modeling of Learning.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Chen, Peter P., Wong, Leah Y., Ting, T. C., and Wong, Leah
- Abstract
The concept of Active Conceptual Modeling of Learning (ACM-L) has been explored in order to capture content and context changes that permit a comprehensive learning from the past, understanding the present, and forecasting the future. Such capability has not been fully explored and it is not available with today's static oriented database system. The potential of creating a "database of intention" that can have its own aim to understand the intentions of its users and the changes to their environment. This paper explores an architectural approach for the "database of intention" with predictability power. The proposed architecture is presented, illustrated, and discussed. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
3. Adaptive and Context-Aware Reconciliation of Reactive and Pro-active Behavior in Evolving Systems.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Chen, Peter P., Wong, Leah Y., Trajcevski, Goce, and Scheuermann, Peter
- Abstract
One distinct characteristics of the context-aware systems is their ability to react and adapt to the evolution of the environment, which is often a result of changes in the values of various (possibly correlated) attributes. Based on these changes, reactive systems typically take corrective actions, e.g., adjusting parameters in order to maintain the desired specifications of the system's state. Pro-active systems, on the other hand, may change the mode of interaction with the environment as well as the desired goals of the system. In this paper we describe our (ECA)2 paradigm for reactive behavior with proactive impact and we present our ongoing work and vision for a system that is capable of context-aware adaptation, while ensuring the maintenance of a set of desired behavioral policies. Our main focus is on developing a formalism that provides tools for expressing normal, as well as defeasible and/or exceptional specification. However, at the same time, we insist on a sound semantics and the capability of answering hypothetical "what-if" queries. Towards this end, we introduce the high-level language $\mathcal L_{\mathcal EAR}$ that can be used to describe the dynamics of the problem domain, specify triggers under the (ECA)2 paradigm, and reason about the consequences of the possible evolutions. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
4. A Common Core for Active Conceptual Modeling for Learning from Surprises.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Chen, Peter P., Wong, Leah Y., Liddle, Stephen W., and Embley, David W.
- Abstract
The new field of active conceptual modeling for learning from surprises (ACM-L) may be helpful in preserving life, protecting property, and improving quality of life. The conceptual modeling community has developed sound theory and practices for conceptual modeling that, if properly applied, could help analysts model and predict more accurately. In particular, we need to associate more semantics with links, and we need fully reified high-level objects and relationships that have a clear, formal underlying semantics that follows a natural, ontological approach. We also need to capture more dynamic aspects in our conceptual models to more accurately model complex, dynamic systems. These concepts already exist, and the theory is well developed; what remains is to link them with the ideas needed to predict system evolution, thus enabling risk assessment and response planning. No single researcher or research group will be able to achieve this ambitious vision alone. As a starting point, we recommend that the nascent ACM-L community agree on a common core model that supports all aspects—static and dynamic—needed for active conceptual modeling in support of learning from surprises. A common core will more likely gain the traction needed to sustain the extended ACM-L research effort that will yield the advertised benefits of learning from surprises. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
5. Actively Evolving Conceptual Models for Mini-World and Run-Time Environment Changes.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Chen, Peter P., Wong, Leah Y., Krishna, P. Radha, and Karlapalem, Kamalakar
- Abstract
Run-time application environments are affected by the changes in mini-world or technology changes. Large number of applications are process driven. For robust applications that can evolve over time, there is a need for a methodology that implicitly handles changes at various levels from mini-world to run-time environment through a layers of models and systems. In this paper, we present ER* methodology for evolving applications. In the context of this paper, the role of two-way active behaviour and template driven development of applications is presented. This methodology facilitates capturing active behaviour from run-time transactions and provides a means of using this knowledge to guide subsequent application design and its evolution. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
6. Achievements and Problems of Conceptual Modelling.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Chen, Peter P., Wong, Leah Y., and Thalheim, Bernhard
- Abstract
Database and information systems technology has substantially changed. Nowadays, content management systems, (information-intensive) web services, collaborating systems, internet databases, OLAP databases etc. have become buzzwords. At the same time, object-relational technology has gained the maturity for being widely applied. Conceptual modelling has not (yet) covered all these novel topics. It has been concentrated for more than two decades around specification of structures. Meanwhile, functionality, interactivity and distribution must be included into conceptual modelling of information systems. Also, some of the open problems that have been already discussed in 1987 [15, 16] still remain to be open. At the same time, novel models such as object-relational models or XML-based models have been developed. They did not overcome all the problems but have been sharpening and extending the variety of open problems. The open problem presented are given for classical areas of database research, i.e., structuring and functionality. The entire are of distribution and interaction is currently an area of very intensive research. The presentation of open problems is combined with the introduction to the achievements of conceptual modelling. The paper develops an approach to conceptual modelling for object-relational, collaborating information systems that support virtual communities of work, integration of information systems, varieties of architecture such as the OLTP-OLAP architecture, varieties of play-out and play-in systems, and data analysis engines. The paper is based on an extended entity-relationship model that covers all structuring facilities of object-relational systems. It uses the theory of media types and storyboards for the specification of interactivity and provides a framework for collaboration. The paper presents 20 open problems that need to be solved for conceptual modelling. The problems are sketched. Main references and the background are given. Additional references can be provided by the author on demand. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
7. Metaphor Modeling on the Semantic Web.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Chen, Peter P., Wong, Leah Y., Czejdo, Bogdan D., Biguenet, Jonathan, and Biguenet, John
- Abstract
Metaphor is a high-level abstract concept that can be an important part of active conceptual modeling. In this paper, we use the extended Unified Modeling Language (UML) for metaphor modeling. We discuss how to create UML diagrams to capture knowledge about metaphors. The metaphor-based processing system on the Semantic Web can support new query/search operations. Such a computer system can be used for a broad spectrum of applications such as predicting surprises (e.g., terrorist attacks) or generating automatically new innovations. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
8. Schema Changes and Historical Information in Conceptual Models in Support of Adaptive Systems.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Chen, Peter P., Wong, Leah Y., Luqi, and Lange, Douglas S.
- Abstract
Conceptual changes and historical information have not been emphasized in traditional approaches to conceptual modeling such as the entity-relationship approach. Effective representations for such changes are needed to support robust machine learning and computer-aided organizational learning. However, these aspects have been modeled and studied in other contexts, such as software maintenance, version control, software transformations, etc. This paper reviews some relevant previous results, shows how they have been used to simplify conceptual models to help people make sense out of complex changing situations, and suggests some connections to conceptual models of machine learning. Areas where research is required to support conceptual models for adaptive systems are also explored. These are suggested by studies of the issues surrounding deployment of adaptive systems in mission critical environments. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
9. Using Active Modeling in Counterterrorism.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Chen, Peter P., Wong, Leah Y., Su, Yi-Jen, Jiau, Hewijin C., and Tsai, Shang-Rong
- Abstract
Terrorist organizations attain their goals by attacking various targets to jeopardize human lives and intimidate governments. As new terrorist attacks almost always aim to break the mold of old plots, tracing the dynamic behaviors of terrorists becomes crucial to national defense. This paper proposes using active modeling in analyzing unconventional attacks in the design of counterterrorism system. The intelligent terrorism detection system not only detects potential threats by monitoring terrorist networks with identified threat patterns, but also continually integrates new threat features in terrorist behaviors and the varying relationships among terrorists. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
10. To Support Emergency Management by Using Active Modeling: A Case of Hurricane Katrina.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Chen, Peter P., Wong, Leah Y., and Xu, Xin
- Abstract
Reducing the complexity and surprise of emergencies can efficiently alleviate negative impacts on society, either local or global. A good model which can simplify complex events by constructing views to describe changes of entity behaviors and the dynamic and time-varying relationships is the key approach to handle surprise crises. Conventional conceptual modeling has limitations on mapping dynamic real world. Active modeling is suggested as the appropriate way to deal with such surprises. A case study of Hurricane Katrina is used as an example to explain some active modeling concepts including temporal conceptual modeling, multi-perspective modeling, and data, information and knowledge integration. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
11. Using Ontological Modeling in a Context-Aware Summarization System to Adapt Text for Mobile Devices.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Chen, Peter P., Wong, Leah Y., Garcia, Luís Fernando Fortes, de Lima, José Valdeni, and Loh, Stanley
- Abstract
This paper presents a context-aware text summarizer based on ontologies intended to be used for adapting information to mobile devices. The system generates summaries from texts according to the profile of the user and the context where he/she is at the moment. Context is determined by spatial and temporal localization. Ontologies are used to allow identifying which parts of the texts are related to the user's profile and to the context. Ontologies are structured as hierarchies of concepts and concepts are represented by keywords with a weight associated. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
12. Accommodating Streams to Support Active Conceptual Modeling of Learning from Surprises.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Chen, Peter P., Wong, Leah Y., and Mazumdar, Subhasish
- Abstract
We argue that a key requirement on an information system that can implement an active conceptual model of learning from surprises is the ability to query data that is not query-able by content, especially data streams;we suggest that such data be queried by context. We propose an enhancement of entity-relationship modeling with active constructs in order to permit such streams to have context-based relationships with standard data. We propose a framework wherein the analysis of surprises and the subsequent monitoring of states that are ripe for such events are possible by the use of such contexts. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
13. Approaches to the Active Conceptual Modelling of Learning.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Chen, Peter P., Wong, Leah Y., and Kangassalo, Hannu
- Abstract
Information modelling is a collection of dynamic processes, in which its content develops from physical processes to abstract knowledge structures. We study that collection on several levels of abstraction of human cognition and knowledge. These processes can be performed through various approaches, on several levels, and by using several perspectives. We concentrate on active conceptual modelling, which has become important in the science and technology, including educational sciences and learning. It is a process of recognition, finding or creating relevant concepts and conceptual models which describe the UoD, representing the conceptual content of information to be contained in the IS. This characterisation contains the construction of new concepts, too. We study methods for collecting information from various sources in the UoD and accumulating it as possibly actual instances of various types of pre-defined concepts. Some of these instances may be cases of sudden events or processes. They should be recognised as concepts and included in to the conceptual schema. To some extent, some concepts may be constructed which fit to this collected information. During the adaptation process we are applying active conceptual modelling for learning, which organises our conceptual schema in a new way. Learning is a process in which a learner re-organises, removes or refills his knowledge structures by applying his newly organised conceptual schema. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
14. Spatio-temporal and Multi-representation Modeling: A Contribution to Active Conceptual Modeling.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Chen, Peter P., Wong, Leah Y., Spaccapietra, Stefano, Parent, Christine, and Zimányi, Esteban
- Abstract
Worldwide globalization increases the complexity of problem solving and decision-making, whatever the endeavor is. This calls for a more accurate and complete understanding of underlying data, processes and events. Data representations have to be as accurate as possible, spanning from the current status of affairs to its past and future statuses, so that it becomes feasible, in particular, to elaborate strategies for the future based on an analysis of past events. Active conceptual modeling is a new framework intended to describe all aspects of a domain. It expands the traditional modeling scope to include, among others, the ability to memorize and use knowledge about the spatial and temporal context of the phenomena of interest, as well as the ability to analyze the same elements under different perspectives. In this paper we show how these advanced modeling features are provided by the MADS conceptual model. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
15. Postponing Schema Definition: Low Instance-to-Entity Ratio (LItER) Modelling.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Chen, Peter P., Wong, Leah Y., Roddick, John F., Ceglar, Aaron, and de Vries, Denise
- Abstract
There are four classes of information system that are not well served by current modelling techniques. First, there are systems for which the number of instances for each entity is relatively low resulting in data definition taking a disproportionate amount of effort. Second, there are systems where the storage of data and the retrieval of information must take priority over the full definition of a schema describing that data. Third, there are those that undergo regular structural change and are thus subject to information loss as a result of changes to the schema's information capacity. Finally, there are those systems where the structure of the information is only partially known or for which there are multiple, perhaps contradictory, competing hypotheses as to the underlying structure. This paper presents the Low Instance-to-Entity Ratio (LItER) Model, which attempts to circumvent some of the problems encountered by these types of application. The two-part LItER modelling process possesses an overarching architecture which provides hypothesis, knowledge base and ontology support together with a common conceptual schema. This allows data to be stored immediately and for a more refined conceptual schema to be developed later. It also facilitates later translation to EER, ORM and UML models and the use of (a form of) SQL. Moreover, an additional benefit of the model is that it provides a partial solution to a number of outstanding issues in current conceptual modelling systems. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
16. Research Issues in Active Conceptual Modeling of Learning: Summary of Panel Discussions in Two Workshops (May 2006) and (November 2006).
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Chen, Peter P., Wong, Leah Y., Delcambre, Lois, Akoka, Jacky, and Sølvberg, Arne
- Abstract
The SPAWAR Systems Center (SSC San Diego) of the U.S. Navy hosted two workshops on Active Conceptual Modeling of Learning (ACM-L). The first workshop was held at SSC San Diego on May 10-12, 2006 to introduce the Science &Technology (S&T) Initiative and identify a Research and Development agenda for the technology development investigation. Eleven invited researchers in Conceptual Modeling presented position papers on the proposed S&T Initiative. The second workshop was held on November 8, 2006 at the 25th International Conference on Conceptual Modeling, ER 2006, 6-9 November 2006, in Tucson, Arizona. Complementary to the May Workshop, the November workshop was a forum for the international researchers and practitioners to present their papers as a result of a call for papers and to exchange ideas from various perspectives of the subject. This paper describes research issues identified by participants from the two ACM-L workshops. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
17. A Novel Approach for Filtering Junk Images from Google Search Results.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Nack, Frank, Etoh, Minoru, Yuli Gao, Jianping Fan, and Hangzai Luo
- Abstract
Keyword-based image search engines such as Google Images are now very popular for accessing large amount of images on the Internet. Because only the text information that are directly or indirectly linked to the images are used for image indexing and retrieval, most existing image search engines such as Google Images may return large amount of junk images which are irrelevant to the given queries. To filter out the junk images from Google Images, we have developed a kernel-based image clustering technique to partition the images returned by Google Images into multiple visually-similar clusters. In addition, users are allowed to input their feedbacks for updating the underlying kernels to achieve more accurate characterization of the diversity of visual similarities between the images. To help users assess the goodness of image kernels and the relevance between the returned images, a novel framework is developed to achieve more intuitive visualization of large amount of returned images according to their visual similarity. Experiments on diverse queries on Google Images have shown that our proposed algorithm can filter out the junk images effectively. Online demo is also released for public evaluation at: http://www.cs.uncc.edu/~jfan/google−demo/. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
18. MILC2: A Multi-Layer Multi-Instance Learning Approach to Video Concept Detection.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Zhiwei Gu, and Tao Mei
- Abstract
Video is a kind of structured data with multi-layer (ML) information, e.g., a shot is consisted of three layers including shot, key-frame, and region. Moreover, multi-instance (MI) relation is embedded along the consecutive layers. Both the ML structure and MI relation are essential for video concept detection. The previous work [5] dealt with ML structure and MI relation by constructing a MLMI kernel in which each layer is assumed to have equal contribution. However, such equal weighting technique cannot well model MI relation or handle ambiguity propagation problem, i.e., the propagation of uncertainty of sub-layer label through multiple layers, as it has been proved that different layers have different contributions to the kernel. In this paper, we propose a novel algorithm named MILC2 (Multi-Layer Multi-Instance Learning with Inter-layer Consistency Constraint.) to tackle the ambiguity propagation problem, in which an inter-layer consistency constraint is explicitly introduced to measure the disagreement of inter-layers, and thus the MI relation is better modeled. This learning task is formulated in a regularization framework with three components including hyper-bag prediction error, inter-layer inconsistency measure, and classifier complexity. We apply the proposed MILC2 to video concept detection over TRECVID 2005 development corpus, and report better performance than both standard Support Vector Machine based and MLMI kernel methods. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
19. An Implicit Active Contour Model for Feature Regions and Lines.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Ho-Ryong Jung, and Moon-Ryul Jung
- Abstract
We present a level-set based implicit active contour method which can detect innermost homogeneous regions which are often considered feature regions or lines depending on the width of the regions. The curve evolution algorithm is derived from optimization of energy defined for the evolving curves. The energy has basically three terms: the first is the energy of the regions inside the curves, the second the energy of the bands inside the curves, and the third the energy of the bands outside the curves. If the band width is small, the total energy is minimized when the evolving curves lie at the boundaries of the innermost homogeneous regions, and the regions inside the curves are considered feature regions. Our method contrasts with the Chan-Vese model, which does not have the notion of innermost homogeneous regions but tries to find the curves such that the regions inside and outside them are both homogeneous. Our model approaches Chan-Vese model as the band width is increased, and is equivalent to Chan-Vese model when the band width is sufficiently large, so that all points inside/outside the curves lie within the bands inside/outside the curves, respectively. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
20. New Approach for Hierarchical Classifier Training and Multi-level Image Annotation.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Nack, Frank, Etoh, Minoru, Jianping Fan, Yuli Gao, and Hangzai Luo
- Abstract
In this paper, we have proposed a novel algorithm to achieve automatic multi-level image annotation by incorporating concept ontology and multi-task learning for hierarchical image classifier training. To achieve more reliable image classifier training in high-dimensional heterogeneous feature space, a new algorithm is proposed by incorporating multiple kernels for diverse image similarity characterization, and a multiple kernel learning algorithm is developed to train the SVM classifiers for the atomic image concepts at the first level of the concept ontology. To enable automatic multi-level image annotation, a novel hierarchical boosting algorithm is proposed by incorporating concept ontology and multi-task learning to achieve hierarchical image classifier training. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
21. Extracting Text Information for Content-Based Video Retrieval.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Lei Xu, and Kongqiao Wang
- Abstract
In this paper we present a novel video text detection and segmentation system. In the detection stage, we utilize edge density feature, pyramid strategy and some weak rules to search for text regions, so that high detection rate can be achieved. Meanwhile, to eliminate the false alarms and improve the precision rate, a multilevel verification strategy is adopted. In the segmentation stage, a precise polarity estimation algorithm is firstly provided. Then, multiple frames containing the same text are integrated to enhance the contrast between text and background. Finally, a novel connected components based binarization algorithm is proposed to improve the recognition rate. Experimental results show the superior performance of the proposed system. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
22. Real-Time Video Surveillance Based on Combining Foreground Extraction and Human Detection.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Hui-Chi Zeng, and Szu-Hao Huang
- Abstract
In this paper, we present an adaptive foreground object extraction algorithm for real-time video surveillance, in conjunction with a human detection technique applied in the extracted foreground regions by using AdaBoost learning algorithm and Histograms of Oriented Gradient (HOG) descriptors. Furthermore, a RANSAC-based temporal tracking algorithm is also applied to refine and trace the detected human windows in order to increase the detection accuracy and reduce the false alarm rate. The traditional background subtraction technique usually cannot work well for situations with lighting variations in the scene. The proposed algorithm employs a two-stage foreground/background classification procedure to perform background subtraction and remove the undesirable subtraction results due to shadow, automatic white balance, and sudden illumination change. Experimental results on some real surveillance video are shown to demonstrate the good performance of the proposed adaptive foreground extraction algorithm under a variety of different environments with lighting variations and human detection system. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
23. Detecting and Clustering Multiple Takes of One Scene.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Bailer, Werner, and Felix Lee
- Abstract
In applications such as video post-production users are confronted with large amounts of redundant unedited raw material, called rushes. Viewing and organizing this material are crucial but time consuming tasks. Typically multiple but slightly different takes of the same scene can be found in the rushes video. We propose a method for detecting and clustering takes of one scene shot from the same or very similar camera positions. It uses a variant of the LCSS algorithm to find matching subsequences in sequences of visual features extracted from the source video. Hierarchical clustering is used to group the takes of one scene. The approach is evaluated in terms of correctly assigned takes using manually annotated ground truth. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
24. An Images-Based 3D Model Retrieval Approach.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Yuehong Wang, and Liu, Rujie
- Abstract
This paper presents an images based 3D model retrieval method in which each model is described by six 2D images. The images are generated by three steps: 1) the model is normalized based on the distribution of the surface normal directions; 2) then, the normalized model is uniformly sampled to generate a number of random points; 3) finally, the random points are projected along six directions to create six images, each of which is described by Zernike moment feature. In the comparison of two models, six images of each model are naturally divided into three pairs, and the similarity between two models is calculated by summing up the distances of all corresponding pairs. The effectiveness of our method is verified by comparative experiments. Meanwhile, high matching speed is achieved, e.g., it takes about 3e-5 seconds to compare two models using a computer with Pentium IV 3.00GHz CPU. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
25. ‘Oh Web Image, Where Art Thou?'.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Ahlers, Dirk, and Boll, Susanne
- Abstract
Web image search today is mostly keyword-based and explores the content surrounding the image. Searching for images related to a certain location quickly shows that Web images typically do not reveal their explicit relation to an actual geographic position. The geographic semantics of Web images are either not available at all or hidden somewhere within the the Web pages' content. Our spatial search engine crawls and identifies Web pages with a spatial relationship. Analysing location-related Web pages, we identify photographs based on content-based image analysis as well as image context. Following the photograph classification, a location-relevance classification process evaluates image context and content against the previously identified address. The results of our experiments show that our approach is a viable method for Web image location assessment. Thus, a large number of potentially geographically-related Web images are unlocked for commercially relevant spatial Web image search. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
26. Complementary Variance Energy for Fingerprint Segmentation.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Hou, Z., and Yau, W. Y.
- Abstract
The paper presents a method for fingerprint segmentation, which executes in two phases. The enrolled image is partitioned into block by block and a preliminary segmentation is firstly carried out based on thresholding the blockwise variance. Afterwhich, the segmentation result is refined by energy information, where the blocks with the most "reliable" energy are selected to polish the preliminary result and yield the final segmentation map. Reliability of the method has been demonstrated using NIST and FVC data sets. Comparison with other methods is presented. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
27. Similarity Search in Multimedia Time Series Data Using Amplitude-Level Features.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Aßfalg, Johannes, and Kriegel, Hans-Peter
- Abstract
Effective similarity search in multi-media time series such as video or audio sequences is important for content-based multi-media retrieval applications. We propose a framework that extracts a sequence of local features from large multi-media time series that reflect the characteristics of the complex structured time series more accurately than global features. In addition, we propose a set of suitable local features that can be derived by our framework. These features are scanned from a time series amplitude-levelwise and are called amplitude-level features. Our experimental evaluation shows that our method models the intuitive similarity of multi-media time series better than existing techniques. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
28. Sound Source Localization with Non-calibrated Microphones.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Kobayashi, Tomoyuki, and Kameda, Yoshinari
- Abstract
We propose a new method for localizing a sound source in a known space with non-calibrated microphones. Our method does not need the accurate positions of the microphones that are required by traditional sound source localization. Our method can make use of wide variety of microphone layout in a large space because it does not need calibration step on installing microphones. After a number of sampling points have been stored in a database, our system can estimate the nearest sampling point of a sound by utilizing the set of time delays of microphone pairs. We conducted a simulation experiment to determine the best microphone layout in order to maximize the accuracy of the localization. We also conducted a preliminary experiment in real environment and obtained promising results. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
29. PriSurv: Privacy Protected Video Surveillance System Using Adaptive Visual Abstraction.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Chinomi, Kenta, and Nitta, Naoko
- Abstract
Recently, video surveillance has received a lot of attention as a technology to realize a secure and safe community. Video surveillance is useful for crime deterrence and investigations, but may cause the invasion of privacy. In this paper, we propose a video surveillance system named PriSurv, which is characterized by visual abstraction. This system protects the privacy of objects in a video by referring to their privacy policies which are determined according to closeness between objects in the video and viewers monitoring it. A prototype of PriSurv is able to protect the privacy adaptively through visual abstraction. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
30. Distribution-Based Similarity for Multi-represented Multimedia Objects.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Kriegel, Hans-Peter, and Kunath, Peter
- Abstract
In modern multimedia databases, objects can be represented by a large variety of feature representations. In order to employ all available information in a best possible way, a joint statement about object similarity must be derived. In this paper, we present a novel technique for multi-represented similarity estimation which is based on probability distributions modeling the connection between the distance value and object similarity. To tune these distribution functions to model the similarity in each representation, we propose a bootstrapping approach maximizing the agreement between the distributions. Thus, we capture the general notion of similarity which is implicitly given by the distance relationships in the available feature representations. Thus, our approach does not need any training examples. In our experimental evaluation, we demonstrate that our new approach offers superior precision and recall compared to standard similarity measures on a real world audio data set. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
31. A Multimodal Input Device for Music Authoring for Children.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Akiyama, Yasushi, and Oore, Sageev
- Abstract
We present a novel interface of a digital music authoring tool that is designed for children. Our system consists of a unique multimodal input device and corresponding platform to support children's music authoring. By departing from the conventional design approaches of existing tools, our system provides a simple and intuitive environment for young children to use. The proposed interface is tested with children in a focus group study, and the results of the study are encouraging in that the children were able to perform some of complicated multitrack recording tasks within a single session that lasted for 15 to 20 minutes. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
32. Free-Shaped Video Collage.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Bo Yang, and Tao Mei
- Abstract
With the explosive growth of multimedia data, video presentation has become an important technology for fast browsing of video content. In this paper, we present a novel video presentation technique called "Free-Shaped Video Collage" (FS-Collage), which is motivated from and built upon our previous work on Video Collage [3]. Video Collage is a kind of static summary which selects the most representative regions-of-interest (ROI) from video and seamlessly arranges them on a synthesized image. Unlike Video Collage in which both the shapes of ROI and final collage are fixed as rectangle, we support arbitrary shapes of ROI and a set of collage templates in FS-Collage. Furthermore, we design three ROI arrangement schemes (i.e., book, diagonal, and spiral) for satisfying different video genres. We formulate the generation of FS-Collage as an energy minimization problem and solve the problem by designing a random sampling process. The experiment results show that our FS-Collage achieves satisfying performance. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
33. Aesthetics-Based Automatic Home Video Skimming System.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Wei-Ting Peng, and Yueh-Hsuan Chiang
- Abstract
In this paper, we propose an automatic home video skimming system based on media aesthetics. Unlike other similar works, the proposed system considers video editing theory and realizes the idea of computational media aesthetics. Given a home video and a incidental background music, this system generates a music video (MV) style skimming video automatically, with consideration of video quality, music tempo, and the editing theory. The background music is analyzed so that visual rhythm caused by shot changes in the skimming video are synchronous with the music tempo. Our work focuses on the rhythm over aesthetic features, which is more recognizable and more suitable to describe the relationship between video and audio. Experiments show that the generated skimming video is effective in representing the original input video, and the audio-video conformity is satisfactory. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
34. Using Fuzzy Lists for Playlist Management.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Deliège, François, and Pedersen, Torben Bach
- Abstract
The increasing popularity of music recommendation systems and the recent growth of online music communities further emphasizes the need for effective playlist management tools able to create, share, and personalize playlists. This paper proposes the development of generic playlists and presents a concrete scenario to illustrate their possibilities. Additionally, to enable the development of playlist management tools, a formal foundation is provided. Therefore, the concept of fuzzy lists is defined and a corresponding algebra is developed. Fuzzy lists offer a solution perfectly suited to meet the demands of playlist management. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
35. Tagging Video Contents with Positive/Negative Interest Based on User's Facial Expression.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Miyahara, Masanori, and Aoki, Masaki
- Abstract
Recently, there are so many videos available for people to choose to watch. To solve this problem, we propose a tagging system for video content based on facial expression that can be used for recommendations based on video content. Viewer's face captured by a camera is extracted by Elastic Bunch Graph Matching, and the facial expression is recognized by Support Vector Machines. The facial expression is classified into Neutral, Positive, Negative and Rejective. Recognition results are recorded as "facial expression tags" in synchronization with video content. Experimental results achieved an averaged recall rate of 87.61%, and averaged precision rate of 88.03%. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
36. Snap2Play: A Mixed-Reality Game Based on Scene Identification.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Tat-Jun Chin, and Yilun You
- Abstract
The ubiquity of camera phones provides a convenient platform to develop immersive mixed-reality games. In this paper we introduce such a game which is loosely based on the popular card game "Memory", where players are asked to match a pair of identical cards among a set of overturned cards by revealing only two cards at a time. In our game, the players are asked to match a "physical card", which is an image of a scene in the real world, to a "digital card", which corresponds to a scene in a virtual world. The objective is to convey a mixed-reality sensation. Cards are matched with a scene identification engine which consists of multiple classifiers trained on previously collected images. We present our comprehensive overall game design, as well as implementation details and results. Additionally, we also describe how we constructed our scene identification engine and its performance. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
37. Real-Time Multi-view Object Tracking in Mediated Environments.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Huan Jin, and Gang Qian
- Abstract
In this paper, we present a robust approach to real-time tracking of multiple objects in mediated environments using a set of calibrated color and IR cameras. Challenges addressed in this paper include robust object tracking in the presence of color projections on the ground plane and partial/complete occlusions. To improve tracking in such complex environment, false candidates introduced by ground plane projection or mismatching of objects between views are removed by using the epipolar constraint and the planar homography. A mixture of Gaussian is learned using the expectation-maximization algorithm for each target to further refine the 3D location estimates. Experimental results demonstrate that the proposed approach is capable of robust and accurate 3D object tracking in a complex environment with a great amount of visual projections and partial/complete occlusions. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
38. Reconstruct 3D Human Motion from Monocular Video Using Motion Library.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Wenzhong Wang, and Xianjie Qiu
- Abstract
In this paper, we present a new approach to reconstruct 3D human motion from video clips with the assistance of a precaputred motion library. Given a monocular video clip recording of one person performing some kind of locomotion and a motion library consisting of similar motions, we can infer the 3D motion from the video clip. We segment the video clip into segments with fixed length, and by using a shape matching method we can find out from the motion library several candidate motion sequences for every video segment, then from these sequences a coarse motion clip is generated by performing a continuity test on the boundaries of these candidate sequences. We propose a pose deformation algorithm to refine the coarse motion. To guarantee the naturalness of the recovered motion, we apply a motion splicing algorithm to the motion clip. We tested the approach using synthetic and real sports videos. The experimental results show the effectiveness of this approach. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
39. Appropriate Segment Extraction from Shots Based on Temporal Patterns of Example Videos.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Kurihara, Yousuke, and Nitta, Naoko
- Abstract
Videos are composed of shots, each of which is recorded continuously by a camera, and video editing can be considered as a process of re-sequencing shots selected from original videos. Shots usually include redundant segments, which are often edited out by professional editors. This paper proposes a method for automatically extracting appropriate segments to be included in edited videos from shots based on temporal patterns of audio and visual features in appropriate segments and redundant segments in example videos learned with Hidden Markov Models. The effectiveness of the proposed method is verified with experiments using original shots extracted from movies and appropriate segments extracted from movie trailers as example videos. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
40. Blurred Image Detection and Classification.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Ping Hsu, and Bing-Yu Chen
- Abstract
Digital photos are massively produced while digital cameras are becoming popular, however, not every photo has good quality. Blur is one of the conventional image quality degradation which is caused by various factors. In this paper, we propose a scheme to detect blurred images and classify them into several different categories. The blur detector uses support vector machines to estimate the blur extent of an image. The blurred images are further classified into either locally or globally blurred images. For globally blurred images, we estimate their point spread functions and classify them into camera shake or out of focus images. For locally blurred images, we find the blurred regions using a segmentation method, and the point spread function estimation on the blurred region can sort out the images with depth of field or moving object. The blur detection and classification processes are fully automatic and can help users to filter out blurred images before importing the photos into their digital photo albums. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
41. Cross-Lingual Retrieval of Identical News Events by Near-Duplicate Video Segment Detection.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Ogawa, Akira, and Takahashi, Tomokazu
- Abstract
Recently, for reusing large quantities of accumulated news video, technology for news topic searching and tracking has become necessary. Moreover, since we need to understand a certain topic from various viewpoints, we focus on identical event detection in various news programs from different countries. Currently, text information is generally used to retrieve news video. However, cross-lingual retrieval is complicated by machine translation performance and different viewpoints and cultures. In this paper, we propose a cross-lingual retrieval method for detecting identical news events that exploits image information together with text information. In an experiment, we verified the effectiveness of making use of the existence of near-duplicate video segments and the possibility of improving retrieval performance. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
42. Web Image Gathering with a Part-Based Object Recognition Method.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, and Yanai, Keiji
- Abstract
We propose a new Web image gathering system which employs a part-based object recognition method. The novelty of our work is introducing the bag-of-keypoints representation into an Web image gathering task instead of color histogram or segmented regions our previous system used. The bag-of-keypoints representation has been proven that it has the excellent ability to represent image concepts in the context of visual object categorization / recognition in spite of its simplicity. Most of object recognition work assumed that complete training data is available. On the other hand, in the Web image gathering task, since images associated with the given keywords are gathered from the Web fully-automatically, complete training images cannot be available. In this paper, we combine the HTML-based automatic positive training image selection and the bag-of-keypoints-based image selection with an SVM which is a supervised machine learning method. This combination enables the system to gather many images related to given concepts with high precision fully automatically needing no human intervention. Our main objective is to examine if the bag-of-keypoints model is also effective for the Web image gathering task where training images always include some noise. By the experiments, we show the new system outperforms our previous systems, other systems and Google Image Search greatly. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
43. A Query Language Combining Object Features and Semantic Events for Surveillance Video Retrieval.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Thi-Lan Le, and Thonnat, Monique
- Abstract
In this paper, we propose a novel query language for video indexing and retrieval that (1) enables to make queries both at the image level and at the semantic level (2) enables the users to define their own scenarios based on semantic events and (3) retrieves videos with both exact matching and similarity matching. For a query language, four main issues must be addressed: data modeling, query formulation, query parsing and query matching. In this paper we focus and give contributions on data modeling, query formulation and query matching. We are currently using color histograms and SIFT features at the image level and 10 types of events at the semantic level. We have tested the proposed query language for the retrieval of surveillance videos of a metro station. In our experiments the database contains more than 200 indexed physical objects and 48 semantic events. The results using different types of queries are promising. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
44. Semantic Quantization of 3D Human Motion Capture Data Through Spatial-Temporal Feature Extraction.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Jin, Yohan, and Prabhakaran, B.
- Abstract
3D motion capture is a form of multimedia data that is widely used in animation and medical fields (such as physical medicine and rehabilitation where body joint analysis is needed). These applications typically create large repositories of motion capture data and need efficient and accurate content-based retrieval techniques. 3D motion capture data is in the form of multi-dimensional time series data. To reduce the dimensions of human motion data while maintaining semantically important features, we quantize human motion data by extracting Spatial-Temporal Features through SVD and translate them onto a 1-dimensional sequential representation through our proposed sGMMEM (semantic Gaussian Mixture Modeling with EM). Thus, we achieve good classification accuracies for primitive human motion categories (walking 92.85%,run 91.42%,jump 94.11%) and even for subtle categories (dance 89.47%,laugh 83.33%,basketball signal 85.71%,golf putting 80.00%). [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
45. Fast Intermode Decision Via Statistical Learning for H.264 Video Coding.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Wei-Hau Pan, and Chen-Kuo Chiang
- Abstract
Although the variable-block-size motion compensation scheme significantly reduces the compensation error, the computational complexity of motion estimation (ME) is tremendously increased at the same time. To reduce the complexity of the variable-block-size ME algorithm, we propose a statistical learning approach to simplify the computation involved in the sub-MB mode selection. Some representative features are extracted during ME with fixed sizes. Then, an off-line pre-classification approach is used to predict the most probable sub-MB modes according to the run-time features. It turns out that only possible sub-MB modes need to perform ME. Experimental results show that the computation complexity is significantly reduced while the video quality degradation and bitrate increment is negligible. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
46. A Novel Motion Estimation Method Based on Normalized Cross Correlation for Video Compression.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Wei, Shou-Der, and Wei-Hau Pan
- Abstract
In this paper we propose to use the normalized cross correlation (NCC) as the similarity measure for block-based motion estimation (ME) to replace the sum of absolute difference (SAD) measure. NCC is a more suitable similarity measure than SAD for reducing the temporal redundancy in video comparison since we can obtain flatter residual after motion compensation by using the NCC as the similarity measure in the motion estimation. The flat residual results in large DC term and smaller AC term, which means less information is lost after quantization. Thus, we can obtain better quality in the compressed video. Experimental results show the proposed NCC-based motion estimation algorithm can provide similar PSNR but better SSIM than the traditional full search ME with the SAD measure. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
47. Curved Ray-Casting for Displacement Mapping in the GPU.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Kyung-Gun Na, and Moon-Ryul Jung
- Abstract
To achieve interactive speed, displacement mapping in the GPU is typically implemented in two steps: vertex shading/rasterization of the base surface and pixel shading. Pixel shading applies the height map relative to the image plane of the base surface, casts view rays to the height field through each pixel, finds the intersection point with the height field, and computes the color of that point. Here, the ray-casting process involves significant errors; The spatial relationship between the ray and the base surface is not preserved between the ray and the image plane of the base surface. The errors result in incorrect silhouettes. To address this problem, we curve the ray so that the spatial relationship between the (linear) ray and the base surface is preserved between the curved ray and the image plane of the base surface. This method reduces intersection errors, producing more satisfactory silhouettes, self-occlusions and shadows. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
48. Emotion-Based Music Visualization Using Photos.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Chin-Han Chen, and Ming-Fang Weng
- Abstract
Music players for personal computers are often featured with music visualization by generating animated patterns according to the music's low-level features such as loudness and spectrum. This paper proposes an emotion-based music player which synchronizes visualization (photos) with music based on the emotions evoked by auditory stimulus of music and visual content of visualization. For emotion detection from photos, we collected 398 photos with their emotions annotated by 496 users through the web. With these annotations, a Bayesian classification method is proposed for automatic photo emotion detection. For emotion detection from music, we adopt an existing method. Finally, for composition of music and photos, in addition to matching high-level emotions, we also consider low-level feature harmony and temporal visual coherence. It is formulated as an optimization problem and solved by a greedy algorithm. Subjective evaluation shows emotion-based music visualization enriches users' listening experiences. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
49. LightCollabo: Distant Collaboration Support System for Manufacturers.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Iyoda, Tetsuo, and Abe, Tsutomu
- Abstract
This paper introduces our LightCollabo system which is a distant collaboration tool designed for manufacturers. For distant collaboration between factory workers and members of a product design team, it is important to support interactive discussion over actual objects. LightCollabo enables remotely located people to share a view of actual objects and to annotate on the objects by using a camera and a projector whose optic axes and view angles are precisely aligned. LightCollabo has been experimentally deployed in several locations in China and Japan and proven to be effective. Initial results of our use case study are also described. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
50. Accurate Identifying Method of JPEG2000 Images for Digital Cinema.
- Author
-
Hutchison, David, Kanade, Takeo, Kittler, Josef, Kleinberg, Jon M., Mattern, Friedemann, Mitchell, John C., Naor, Moni, Nierstrasz, Oscar, Pandu Rangan, C., Steffen, Bernhard, Sudan, Madhu, Terzopoulos, Demetri, Tygar, Doug, Vardi, Moshe Y., Weikum, Gerhard, Satoh, Shin'ichi, Nack, Frank, Etoh, Minoru, Fukuhara, Takahiro, and Hosaka, Kazuhisa
- Abstract
JPEG2000 has been selected as the image compression and decompression technology for digital cinema by DCI(Digital Cinema Initiatives). In a few years, large amount of JPEG2000 compressed files will be widely used and delivered. Identifying a target compressed file out of image sequences or database is highly demanded. In this paper, very unique identifying method is presented. It takes advantage of the feature of JPEG2000 codestream structure. The proposed method has advantages of low computation and high accuracy over conventional methods. [ABSTRACT FROM AUTHOR]
- Published
- 2008
- Full Text
- View/download PDF
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.