Papers and Tools Download

E-learning

Link to E-learning System

Video Processing

Link to Video Professing

Interactive Multimedia and Music

Link to Interactive Multimedia and music

 

Selected Papers

Nick C. Tang, Chiou-Ting Hsu, Chih-Wen Su, Timothy K. Shih and Hong-Yuan Mark Liao, “Video Inpainting on Digitized Vintage Films via Maintaining Spatiotemporal Continuity,” in IEEE Transactions on Multimedia, Vol. 13, No. 4, August 2011.

 ABSTRACT

In line with the popularity of the Internet and the development of search engine, users request information through web-based services. Although general-purpose searching such as one provided by Google is powerful, searching mechanism for specific purposes could rely on metadata. In distance learning (or e-learning), SCORM provides an efficient metadata definition for learning objects to be searched and shared. To facilitate searching in a federated repository, CORDRA provides a common architecture for discovering and sharing Learning Objects. We followed SCORM and CORDRA specifications to develop a registry system, called the MINE Registry, for storing and sharing 20,738 Learning Objects created in the past five years. As a contribution, we propose the concept of “Reusability Tree” to represent the relationships among relevant Learning Objects and enhance CORDRA. We further collect relevant information, while users are utilizing Learnin Objects, such as citations and time period persisted. The feedbacks from the user community are also considered as critical elements for evaluating significance degree of Learning Objects. Through these factors, we propose a mechanism to weight and rank Learning Objects in the MINE Registry, in addition to other external learning objects repositories. As a practical contribution, we provide a tool called “Search Guider” to assist users in finding relevant information in Learning Objects based on individual requirements.

 Index Terms—Frame completion, motion completion, motion estimation, video inpainting.

 

Link to PDF

 

Neil Y. Yen, Timothy K. Shih, Louis R. Chao, Qun Jin, “Ranking Metrics and Search Guidance for Learning Object Repository,” in IEEE Transactions on Learning Technologies, Vol. 3, No. 3, July-September 2010.

ABSTRACT

In line with the popularity of the Internet and the development of search engine, users request information through web-based services. Although general-purpose searching such as one provided by Google is powerful, searching mechanism for specific purposes could rely on metadata. In distance learning (or e-learning), SCORM provides an efficient metadata definition for learning objects to be searched and shared. To facilitate searching in a federated repository, CORDRA provides a common architecture for discovering and sharing Learning Objects. We followed SCORM and CORDRA specifications to develop a registry system, called the MINE Registry, for storing and sharing 20,738 Learning Objects created in the past five years. As a contribution, we propose the concept of “Reusability Tree” to represent the relationships among relevant Learning Objects and enhance CORDRA. We further collect relevant information, while users are utilizing Learnin Objects, such as citations and time period persisted. The feedbacks from the user community are also considered as critical elements for evaluating significance degree of Learning Objects. Through these factors, we propose a mechanism to weight and rank Learning Objects in the MINE Registry, in addition to other external learning objects repositories. As a practical contribution, we provide a tool called “Search Guider” to assist users in finding relevant information in Learning Objects based on individual requirements.

 

Link to PDF

 

Timothy K. Shih, Nick Tang, Joseph Tsai, Jenq-Neng Hwang, “Video Motion Interpolation for Special Effect Applications,” in IEEE Transactions on Systems, Man, and Cybernetics –Part C: Applications and Reviews, Vol. 41, No. 5, September 2011.

 ABSTRACT

Video  forgery,  also  referred  as  video  falsifying,  is a  technique  for  generating  fake  videos  by  altering,  combining, or creating new video contents. For instance, the outcome of a 100 m race in the olympic game is forged as an example in this paper. We track objects and segment motions using a modified mean shift mechanism. The resulting video layers can be played in different speeds and from different reference points with re- spect to the original video. In order to obtain a smooth move- ment of target objects, a motion interpolation mechanism is pro-posed based on reference stick figures (i.e., a structure of human skeleton) and a video inpainting mechanism. The video inpaint-ing mechanism is performed in a quasi-3-D space via guided 3-D patch matching. Interpolated target objects and background layers are then fused. The objective is to create a forged video, which is almost indistinguishable from the original video. We demonstrate the original and the forged videos in our Web site at http://member.mine.tku.edu.tw/www/TSMC09/. Although video forgery may create moral or legal issues, which is beyond the scope of this paper, our intension is to create special effects in video editing applications.

 

Index Terms—Image completion, mean shift, motion interpola-tion, object tracking, video falsifying, video forgery, video inpaint-ing, video special effect.

 

Link to PDF

 

  Freya H. Lin and Timothy K. Shih, “Automatic Trap Detection: A Debugging Mechanism for Abnormal Specification in the IMS Sequencing Controls,” in IEEE Transactions on Learning Technologies, Vol. 1, No. 3, Jul-September 2008.

 ABSTRACT

Adaptive learning is a challenging trend of distance education. In order to adapt diversified teaching in accordance with individual abilities, educators emphasize on constructing personalized courseware as an Internet-based application. To cope with the drawback of nonsupervised blind spots on the dynamic navigation of learners, the SCORM 2004 specification adds the concept of sequences (i.e., based on IMS Simple Sequence Specification) into its content delivery. Sequencing rules describe how individual navigation is invoked, what learning objects are delivered during the navigation process, and the potential outputs of the process. However, there is a potential leak from using sequence specification. It is possible for an instruction designer to make a mistake similar to that made by any programmer in software development. In other words, a sequencing trap results from improper sequence composition, which may block learning activities from a normal navigation. Although the problem can be manually checked with an ordinary SCORM presentation tool, nonsystematic debugging can be an unsound solution and a time consuming job, since the complexity of courseware is usually proportional to the occurrences of traps. We point out the sequencing trap problem and its solution by using a fully automatic mechanism for trap detection. A Petri net is used in the computation of sequencing traps. The contribution of this paper can be used as a first technique for automatic sequencing testing in the IMS Simple Sequence Specification and SCORM 2004.

 

Index Terms—IMS Simple Sequence Specification, SCORM 2004, Petri net, automatic testing, distance learning, Web-based learning, sequence debugging.

 

Link to PDF

 

 Freya H. Lin, Timothy K. Shih, and Won Kim, “An Implementation of the CORDRA Architecture Enhanced for Systematic Reuse of Learning Objects,” IEEE Transactions on Knowledge and Data Engineering, Vol. 21, No. 6, June 2009.

 ABSTRACT

The Sharable Content Object Reference Model (SCORM) specification defines metadata of learning objects, which are used as the elementary reusable components in distance learning. The Content Object Repository Discovery and Registration/Resolution Architecture (CORDRA) specification provides a common architecture for the resolution, discovery, and sharing of these learning objects. These two specifications together define standardized ways in which learning objects can be discovered and reused by content designers. However, the current CORDRA and the definition of objects in SCORM only allow an object to be copied, updated, and reorganized in a new content aggregation, which is used as a delivery package to end users. This paper proposes a revised CORDRA architecture and a reusability mechanism to make instruction design easier. In particular, it proposes a structure called a reusability tree for tracking the history of reuse of learning objects in CORDRA. This paper also defines the notions of similarity, diversity, and relevancy of learning objects to make it easier for users to precisely search for and reuse learning objects.

 

Index Terms—SCORM, CORDRA, reusability, learning object metadata, distributed system, distance learning.

 

Link to PDF

 

 Timothy K. Shih, Nick C. Tang and Jenq-Neng Hwang, “Exemplar-based Video Inpainting without Ghost Shadow Artifacts by Maintaining Temporal Continuity,” in IEEE Transactions on Circuits and Systems for Video Technology, Vol. 19, No. 3, March 2009.

 ABSTRACT

 Image inpainting or image completion is  the  technique that automatically restores/completes removed areas in an image. When dealing with a similar problem in video, not only should a robust tracking algorithm be used, but the temporal con- tinuity among video frames also needs to be taken into account, especially when the video has camera motions such as zooming and tilting. In this paper, we extend an exemplar-based image inpainting algorithm by incorporating an improved patch matching strategy for video inpainting. In our proposed algorithm, different motion segments with different temporal continuity call for different candidate patches, which are used to inpaint holes after a selected video object is tracked and removed. The proposed new video inpainting algorithm produces very few “ghost shadows,” which were produced by  most  image  inpainting  algorithms  di- rectly applied on video. Our experiments use different types of videos, including cartoon, video from games, and video from digital camera with different camera motions. Our demonstration at http://member.mine.tku.edu.tw/www/T CSVT/web/shows the promising results.

 

Index Terms—Digital inpainting, image completion, motion map segmentation, object removal, object tracking, video inpainting, video special effect.

 

Link to PDF

 

 Timothy K. Shih, Te-Hua Wang, Chih-Yung Chang, Tai-Chien Kao and Douglas Hamilton, “Ubiquitous e-Learning with Multimodal Multimedia Devices,” in IEEE Transactions on Multimedia, Vol. 9, No. 3, April 2007.

 ABSTRACT

 The   Sharable   Content   Object   Reference   Model (SCORM) is a set of specifications and guidelines for the repre-sentation and operation of asynchronous distance learning. Since it  was  announced  in  late  1990s,  the  reference  model  has  been used by software developers and academics in the development of authoring tools, learning management systems, and repositories for distance  learning content. To date, most  e-learning systems have  been  based  on  multimedia  and  Web  technologies  on  per-sonal computers. Our project, Hard SCORM, advances the field by implementing an integrated system which allows learners to read SCORM-compliant textbooks using multimodal multimedia devices. Hard SCORM employs a pen-like optical character reader device (called Hyper Pen) as an input mechanism. A computer, a personal digital assistant, or a cellular phone can be used for user behavior supervision using the Hard SCORM Machine. With an au-thoring tool, specially designed tags are printed in textbooks and recognized by Hyper Pen for user navigation control. In this way, users can read hardcopy textbooks in a traditional manner while the process of reading conforms to the SCORM specification. Part of the implemented system (Pocket SCORM on PDA) received the 2005 Brandon Hall Excellence in Learning Awards. The system has also been used by an airline company for online security checking and a high school for online mobile learning.

 

Index Terms—Augmented Paper, distance Learning, IEEE LOM, mobile devices, SCORM, web service.

 

Link to PDF

 

  Jihoon Myung, Wonjun Lee, Jaideep Srivastava, and Timothy K. Shih, “Tag-Splitting: Adaptive Collision Arbitration Protocols for RFID Tag Identification,” in IEEE Transactions on Parallel and Distributed Systems, Vol. 18, No. 6, June 2007

ABSTRACT

 Tag identification is an important tool in RFID systems with applications for monitoring and tracking. A RFID reader recognizes tags through communication over a shared wireless channel. When multiple tags transmit their IDs simultaneously, the tag-to-reader signals collide and this collision disturbs a reader’s identification process. Therefore, tag collision arbitration for passive tags is a significant issue for fast identification. This paper presents two adaptive tag anticollision protocols: an Adaptive Query Splitting protocol (AQS), which is an improvement on the query tree protocol, and an Adaptive Binary Splitting protocol (ABS), which is based on the binary tree protocol and is a de facto standard for RFID anticollision protocols. To reduce collisions and identify tags efficiently, adaptive tag anticollision protocols use information obtained from the last process of tag identification. Our performance evaluation shows that AQS and ABS outperform other tree-based tag anticollision protocols.

 

Index Terms—Collision resolution, RFID, tag anticollision, tag identification.

 

Link to PDF

 

   Jihoon Myung, Wonjun Lee, and Timothy K. Shih, “An Adaptive Memoryless Protocol for RFID Tag Collision Arbitration,” in IEEE Transactions on Multimedia, Vol. 8, No. 5, October 2006.

ABSTRACT.

 A radio frequency identification (RFID) reader recognizes objects through wireless communications with RFID tags. Tag collision arbitration for passive tags is a significant issue for fast tag identification due o communication over a shared wireless channel. This paper presents an daptive memoryless protocol, which is an improvement on the query tree tocol. Memoryless means that tags need not have additional memory except ID for identification. To reduce collisions and identify tags promptly, we use information obtained from the last process of tag identification at a reader. Our performance evaluation shows that the adaptive memoryless protocol causes fewer collisions and takes shorter delay for recognizing all tags while preserving lower communication overhead than other tree based tag anticollision protocols.

 

Index TermsCollision resolution, RFID, tag anticollision, tag identifi-cation.

 

Link to PDF