• DocumentCode
    3537232
  • Title

    Extracting the Path of Frame Center Points Using Spatial Transformation and Motion Estimation

  • Author

    Dinh, Toan Nguyen ; Kim, JeongHwan ; Lee, Gueesang

  • Author_Institution
    Dept. of Electron. & Comput. Eng., Chonnam Nat. Univ., Gwangju, South Korea
  • fYear
    2011
  • fDate
    Aug. 31 2011-Sept. 2 2011
  • Firstpage
    283
  • Lastpage
    290
  • Abstract
    A new interactive technique for mobile phones using built-in cameras is presented. Users can use the path that connects center points of frames in the captured image sequence to indicate an object contour or a curve text line. To estimate the camera movement, Speeded-Up Robust Features (SURF) is used to estimate the spatial transformation parameters between the previous and current frames. These parameters are then used to recalculate the locations of the center points in the previous frame into the current frame. Finally, the motion estimation with a small search range is applied to refine the locations of the center points. The experimental results obtained from real image sequences show that the proposed method can generate accurate and stable results.
  • Keywords
    cameras; image sequences; mobile handsets; motion estimation; parameter estimation; built-in cameras; camera movement estimation; curve text line; frame center points; image sequence; interactive technique; mobile phones; motion estimation; object contour; path extraction; spatial transformation; spatial transformation parameter estimation; speeded-up robust features; Cameras; Feature extraction; Hidden Markov models; Image sequences; Mathematical model; Mobile handsets; Motion estimation; SURF; global motion; image sequence; mobile phone camera; motion estimation; spatial transformation;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer and Information Technology (CIT), 2011 IEEE 11th International Conference on
  • Conference_Location
    Pafos
  • Print_ISBN
    978-1-4577-0383-6
  • Electronic_ISBN
    978-0-7695-4388-8
  • Type

    conf

  • DOI
    10.1109/CIT.2011.53
  • Filename
    6036773