Computer Science

@

Dr. Ramazan S. Aygün

Developing Novel Mosaic Generation Methods
for Object-Based Multimedia Information Systems

 

 

 

 

 

 Project Home

 People

 Progress

 Publications

 Thesis/Dissertation

 Announcement

 IEEE Video Panorama  Workshop

 K12 Workshop

 Other Projects

 Aygün Home

 

  • Synthetic video generation
    • Used our Mtool(the tool to generate synthetic videos by applying camera motion patterns)to generate large object test cases.(Dec,2010)
    • The paper "Synthetic Video Generation for Evaluation of Sprite Generation" got accepted.
    • Gave presentations including one demo presentation on IEEE International Symposium on Multimedia.(Dec, 2009)
    • Worked on the response of "Synthetic Video Generation for Evaluation of Sprite Generation"(Nov, 2009-Jan, 2010)
    • Submitted the paper titled "Synthetic Video Generation for Evaluation of Sprite Generation" to The International Journal of Multimedia Data Engineering and Management. (Oct-Nov, 2009)
    • Worked on object mask for sprite generation for tracking boat video and tracking field video.(Sep, 2009- )
    • Built a web-based tool for sprite generation and submitted the paper titled "SpriteDB tool – a web-tool for comparing sprite generation results"to The IEEE International Symposium on Multimedia (July 15-Aug 20, 2009)
    • Experimented on generated synthetic sequence videos from 7 testfiles by sprite generation on the complex camera motion patterns. Submitted a paper titled "Synthetic Video Generation with Complex Camera Motion Patterns to Evaluate Sprite Generation" to The Fifth IEEE International Workshop on Multimedia Information Processing and Retrieval.(July 15-Aug 20, 2009)
    • Designed and implemented SpriteDB web-based tool.(May1-July 1, 2009)
    • Implemented more complex patterns including rotate, Affine, PTZ and combined pattern. (May 2009-July, 2009)
    • Experimented on generated synthetic sequence videos from 7 testfiles by sprite generation.(Feb 29-Mar , 2009). Test result is also statisticalled analyized and submitted a paper titled "Synthetic Video Generation with Camera Motion Patterns to Evaluate Sprite Generation" to The First International Conference on Advances in Multimedia MMEDIA 2009 (Feb 29-Mar 14, 2009).
    • Designed and implementated zigzag, spiral and etc pattern to generate synthetic sequence video.(Jan 26-Mar 7, 2009).
  • Sprite Generation
    • Eliminated insignificant macroblocks in the global motion estimation.(June, 2011-Aug, 2011)
    • Obtained permissions of videos for sprite from IEEE, Ebru TV, and Athletic Management System Corporation for the paper.(Feb, 2011- Jun, 2011)
    • The paper titled "Sprite Generation Using Sprite Fusion" got accepted by ACM Transactions on Multimedia Computing, Communications and Applications.(Feb, 2011)
    • Improved the algorithm by applying fusion step as early as possible.(Dec, 2010-Jan, 2011)
    • Computed our Sprite fusion method with two algorithms to show the efficiency and reliability of our algorithm.(Aug, 2010-Jan, 2011)
    • Adapted large object test case to show the capability of our sprite generation algorithm to estimate global motion parameters correctly.(Aug, 2010- Jan, 2011)
    • Worked on the response of "Sprite Generation Using Sprite Fusion."(Aug, 2010- Jan, 2011)
    • Submitted a paper on sprite fusion method to ACM Transactions on Multimedia Computing, Communications and Applications.(Aug, 2010)
    • Integrated the new classification and test data to website.(Aug 21, 2010)
    • Generated synthetic video with object to testfy the new blending method.(Aug 11-Aug 27, 2010)
    • Worked on alternate sprite generation methods.(Mar 1-Aug, 2010)
  • Interactive Video Reproduction
    • Submitted a paper "SpriteCam: Virtual Camera Control using Sprite" to Journal IET Vision 2011.(June, 2011)
    • Worked on a paper on Interactive Video Reproduction.(April, 2011- June, 2011)
    • Provided 5 test cases to display the functionality of our application.(May, 2011)
    • Provided Aspect Ratio Conversion without distortion by using missing area from the mosaic.(April, 2011)
    • Worked on Manual mode. In manual mode, the users can begin from any frame and begin regenerate the new frame according their favors.(March, 2011)
    • Worked on Automatic mode. In Automatic mode, the camera is statically situated in one location of sprite during capture when the original frames are played on the sprite.(March, 2011)
    • Provided two modes: Automatic and Manual.(March, 2011)
    • Regenerated video using the virtual camera interface.(Feb, 2011)
    • Provided a virtual camera interface to conduct pan, tilit, zoom and rotate based on the original frame. (Feb, 2011)
    • Developed a proper embedding of frames onto the mosaic with correct alignment.(Dec,2010-Jan,2011)
    • Worked on image intepolation problem.(Aug, 2010 -Dec,2010)
    • Solved object-centerlized problem.(Jun, 2010)
    • Replace the sprite with newly generated sprite using fusion technology.(May-Jun, 2010)
    • Finshed different mode fixing.(Mar.1-May, 2010)
  • Interactive retrieval of spatio-temporal data using Gamepad
    • Worked on the journal "GStar:Spatio-Temporal querying and Retrieval Using Gamepad." (April, 2011)
    • The system is tested by users and compared with mouse-based system. (April, 2011)
    • The paper"A User Interface for Spatio-Temporal‘Eventually’ Queries using Gamepad." got accepted by MMedia 2011.
    • Submitted a paper"A User Interface for Spatio-Temporal ‘Eventually’ Queries using Gamepad." to MMedia 2011(Nov, 2010)
    • Improved the user interface.(Jan, 2011)
    • Developed "Eventually" Queries using Gamepad.(Aug, 2010-Feb, 2011)
    • Developed "Next" Query using Gamepad.(Aug, 2010-Feb, 2011)
    • Implemented rough player tracking algorithm. (Aug, 2010-Dec, 2010)
    • In the process of preparing the paper on the work done in interactive retrieval of spatio-temporal data for "IEEE multimedia magazine" (July-Aug, 2010).
    • Experimented the system to verify that relevant clips are retrieved (Jan-Jun, 2010).
    • Worked out on the user interface to make the process of query building more interactive (Jan-Jun, 2010).
    • Proposed a technique using joystick as a device in building spatio-temporal queries. Worked out in mapping signals obtained from joystick to semantic information of the video (Jan-Jun, 2010).
    • Studied literatures towards various techniques proposed in building spatio-temporal queries and interactively retrieving data that change over space and time (Nov, 2009).
    • Analyzed some of the limitation of these systems in retrieving spatio-temporal data from the videos (Oct, 2009).
    • Studied SMART system and semantic sequence state graph (S3G) indexing structure (Aug-Sep, 2009).
  • Video classification
    • Integrated all the video classification tools and designed an automatic video classification tool. (Aug-Dec, 2009)
    • Completed all documentation and defended the thesis titled “Video classification for sprite generation” (Sep, 2009)
    • Publishing an open-source server-side database to display classification result. (May-July, 2009).
    • Studied the DTREG classification tool and compared the improved classification results with the results obtained from DTREG tool. Completed the research by analyzing the classification results and defined the color coding based on the suitability of the video for sprite generation.(May-Jun, 2009)
    • Submitted a paper titled " Motion Based Video Classification for sprite generation" based on the improved classification approach to 3rd International Workshop on Multimedia Data Mining and Management" (May, 2009)
    • Improved the initial classification approach. Introduced a refined feature set and built a new classification with 6 pre-defined classes. Performed experiments on all the videos in the dataset and achieved accuracy of 82%. Studied the effect of varying number of frames being processed for classifying the video. (Apr-Jul, 2009)
    • Submitted a paper titled "Video Classification for sprite generation in dynamic video encoder" based on the initial classification approach to 2009 IEEE International Conference on Image Processing (Mar, 2009).
    • Performed experiments on the collected dataset. Initial classification approach achieved an accuracy of 72%.(Mar, 2009– Apr, 2009)
    • Implemented the feature extraction process and collected the values for all the features defined in the feature set. Designed and built classification tree with 9 pre-defined classes based on the values of the features. (Feb-Apr, 2009)
    • Studied various motion based features / parameters based on local and global motion in the video. Designed the feature set for classifying videos suitable for sprite generation. (Jan – Feb, 2009)
    • Studied various motion based features / parameters based on local and global motion in the video. Designed the feature set for classifying videos suitable for sprite generation. (Jan – Feb, 2009)
    • Collected the video dataset consisting of wide variety of videos such as news, sports, education, documentary etc. (Aug, 2008 – Dec, 2008).
  • Spatial Browsing
    • Working on a new indexing for spatial browsing of a video based on the layout map of a floor in the building.(June,2011- )
  • IEEE Workshop on Video Panorama
    • Paper notification.(Aug, 2011)
    • Paper reviews.(Aug, 2011)
    • Paper submission. (July, 2011- Aug, 2011)
    • Invited committee members.(Feb, 2011-April,2011)
  • Workshop on K12
    • Worked on legal documents for workshop.(May, 2011)
    • Prepared website for the workshop.(May, 2011)

 

The views, opinions, and conclusions expressed in this page are those of the author or organization and not necessarily those of The University of Alabama in Huntsville or its officers and trustees. The content of this page has not been reviewed or approved by UAHuntsville and the author or organization is solely responsible for its content.