-
[show abstract]
[hide abstract]
ABSTRACT: In the daily life, people often repeat regular routes in certain periods. In this paper, a mining system is developed to find
the continuous route patterns of personal past trips. In order to count the diversity of personal moving status, the mining
system employs the adaptive GPS data recording and five data filters to guarantee the clean trips data. The mining system
uses a client/server architecture to protect personal privacy and to reduce the computational load. The server conducts the
main mining procedure but with insufficient information to recover real personal routes. In order to improve the scalability
of sequential pattern mining, a novel pattern mining algorithm, continuous route pattern mining (CRPM), is proposed. This
algorithm can tolerate the different disturbances in real routes and extract the frequent patterns. Experimental results based
on nine persons’ trips show that CRPM can extract more than two times longer route patterns than the traditional route pattern
mining algorithms.
Journal of Zhejiang University - Science A: Applied Physics & Engineering 01/2009; 10(2):221-231. · 0.41 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Distributed Interactive Applications (DIAs) is different with other network-based applications, since the data should be delivered to the receiver node according to its priority. Among the existing overlay multicast protocols the "priority-based directed minimum spanning tree" (PST) is designed for DIAs. But in PST the available bandwidth (avail-bw) is used inefficiently and the system becomes unstable when its scale increases. This paper proposed a novel overlay multicast protocol named Priority Based Overlay Multicast (POM), which uses the entities priority, links delay and nodes avail-bw to build multicast tree. By using POM the nodes with high priority receive the data in short delay, while the nodes with low priority tolerate a long delay in order to balance the available bandwidth usage among the nodes. The simulation results show that POM uses the avail-bw efficiently and keeps the system stable when its scale increases. Keywords: DIAs; Priority; Available bandwidth.
Computers and Communications, 2006. ISCC '06. Proceedings. 11th IEEE Symposium on; 07/2006
-
[show abstract]
[hide abstract]
ABSTRACT: In this paper, an emotion recognition system for mandarin speech is presented. Five basic human emotions including angry,
fear, happy, neutral and sad are investigated. The recognizer is based on neural network with OCON and ACON architecture.
Some novel feature selection methods are also added as optional tool to enhance the efficiency and classification accuracy.
The system can train speaker dependent emotion speech model through online emotional utterance recording. Experiment results
show that emotion can be recognized through neural network model, the best mean accuracy is 86.7%. In addition, the feature
selection module is effective to reduce the compute load and increase the generalization ability of the recognizer.
06/2006: pages 941-948;
-
[show abstract]
[hide abstract]
ABSTRACT: 3D face tracking is useful in many applications such as face recognition, analysis of facial expressions and entertainment. Most existing 3D face tracking methods employ multiple cameras and all cameras need to be calibrated first before tracking. The complexity of these methods restricts the applications of them. This paper presents a new simple face tracking method which is based on a stereo vision camera system. All cameras in this stereo vision camera system are pre-calibrated and the system can provide accurate depth information of the scene. The proposed tracking method adopts KLT and template matching to track 2D facial features. Then the stereo vision camera system is used to obtain corresponding depth information of these features. In order to achieve robust tracking, some spatial constraints of these features are employed. Experiment results indicate that the proposed method can achieve robust and accurate 3D face tracking.
Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on; 09/2005
-
[show abstract]
[hide abstract]
ABSTRACT: An efficient DDM (data distribution management) approach-fuzzy grouping mechanism was presented. In order to reduce the matching times and increase the data filtering rate, this new mechanism adopted two techniques based on the fuzzy correlation space approach: fuzzy consistent relation based weight distribution and grid based preprocessing. Simulation results indicate that: compared with existing approaches, fuzzy grouping mechanism not only keeps high and stable data filtering rates in different situations but also greatly reduces matching times.
Communications, 2005. ICC 2005. 2005 IEEE International Conference on; 06/2005
-
[show abstract]
[hide abstract]
ABSTRACT: Consistency maintenance is a kernel problem in Collaborative Virtual Environment (CVE) research. The approaches used in Networked Virtual Environments (e.g., DR algorithm) could not be used in CVEs, for they could not prevent short-term inconsistency. Therefore, local-lag mechanism has been proposed to eliminate short-term inconsistency in CVEs. Choosing a proper lag value is a key problem in local-lag mechanism. This paper studied the effects of lag value (0ms–900ms) on task performance in a desktop CVE system. Experimental results indicate that the effect of lag value on task performance is not linear. The effect could be separated into four segments by three dividing points: 150ms, 300ms and 600ms. Lag has no effect on task performance while ranging from 0ms to 150ms. From 150ms to 300ms, lag slightly affects task performance. Lag deteriorates task performance seriously while ranging from 300ms to 600ms. When lag is longer than 600ms, the task cannot be accomplished sometimes.
Journal of Computer Science and Technology 01/2005; 20(3):396-401. · 0.56 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: With the appearance of different kinds of networked virtual environment systems, the technology of realistic video avatar has become a hotspot of research. How to create realistic video avatar with less hardware investment and limited computation is one of the purposes of these researches. We introduce a method that combines a single stereo camera system and a 3D head model to create a realistic video avatar. To create the video avatar's most important part-the face, we use a 2D image and a depth map, which is generated by the stereo camera system. The other parts of of the video avatar are created by mapping the head's texture to the 3D head model.
Machine Learning and Cybernetics, 2002. Proceedings. 2002 International Conference on; 02/2002
-
[show abstract]
[hide abstract]
ABSTRACT: Parallel and distributed IR holds great potential for tackling the
performance and scale issues associated with the large and growing
document collections currently available online. We surveyed a number of
techniques for exploiting modern parallel architectures. We can easily
implement a parallel system that works on both multiprocessor and
distributed architectures with relatively little modification. We also
present some challenges on parallel and distributed IR, and the future
work for tackling these issues
Info-tech and Info-net, 2001. Proceedings. ICII 2001 - Beijing. 2001 International Conferences on; 02/2001
-
[show abstract]
[hide abstract]
ABSTRACT: The Web-based real-time cooperative edit system is an efficient working paradigm of CSCW, and it offers a function of editing a document at the same time by multiusers through a Web browser. Modeling versions is one of the most important technologies of the system. According to the characteristics of cooperative work, we define the version, world, context, entity, current version and relationship to modeling version. Then four kinds of meetings are presented, and the different characteristics of their record versions are discussed. The multi-version, dynamic and inter-related properties of the versions in meetings are analyzed. According to the system's version model, an instance is given.
Info-tech and Info-net, 2001. Proceedings. ICII 2001 - Beijing. 2001 International Conferences on; 02/2001
-
[show abstract]
[hide abstract]
ABSTRACT: A char-based consistency maintenance algorithm is introduced which is implemented in the brainstorm edit mode on the Web-based real-time cooperative edit system. Here we put forward a star-like system architecture, which is based on events' intercepting-transmitting-recurring, and uses an improved transformation algorithm that is based on the GOTO algorithm to realize the consistency maintenance. With this improved algorithm, we get high responsiveness of local operations and good consistency among distributed users' operations. We introduce our experimental system WRCES.
Info-tech and Info-net, 2001. Proceedings. ICII 2001 - Beijing. 2001 International Conferences on; 02/2001
-
[show abstract]
[hide abstract]
ABSTRACT: Performance of speech emotion recognition largely depends on the acoustic features used in a classifier. This paper studies
the statistical feature selection problem in Mandarin speech emotion recognition. This study was based on a speaker dependent
emotional mandarin database. Pitch, energy, duration, formant related features and some velocity information were selected
as base features. Some statistics of them consisted of original feature set and full stepwise discriminant analysis (SDA)
was employed to select extracted features. The results of feature selection were evaluated through a LDA based classifier.
Experiment results indicate that pitch, log energy, speed and 1st formant are the most important factors and the accuracy
rate increases from 63.1 % to 76.5 % after feature selection. Meanwhile, the features selected by SDA are better than the
results of other feature selection methods in a LDA based classifier and SVM. The best performance is achieved when the feature
number is in the range of 9 to 12.
01/1970: pages 591-600;
-
[show abstract]
[hide abstract]
ABSTRACT: The development of multimedia annotation technique provides the possibility to redesign the interfaces of widely used media players, and EmoPlayer is such a media player that can be used to play video clips with affective annotations. A user can select a character in a video clip and view the distribution of his/her emotions along the video timeline through a colour bar based interface. Two experiments were conducted to evaluate the efficiency of affective annotation. The results of these experiments indicate that affective annotation is effective in both improving the speed of locating a specific scene within a video clip and helping comprehend a video clip in a limited viewing time period. Based on the analysis of recorded operations of participants, the strategies employed by participants and the factors that might influence the utilization of affective annotation are also highlighted.
Interacting with Computers.
