National Tsing Hua University Institutional Repository:Video object inpainting using manifold-based posture estimation
English  |  正體中文  |  简体中文  |  全文筆數/總筆數 : 54367/62174 (87%)
造訪人次 : 14544177      線上人數 : 71
RC Version 6.0 © Powered By DSPACE, MIT. Enhanced by NTHU Library IR team.
搜尋範圍 查詢小技巧:
  • 您可在西文檢索詞彙前後加上"雙引號",以獲取較精準的檢索結果
  • 若欲以作者姓名搜尋,建議至進階搜尋限定作者欄位,可獲得較完整資料
  • 進階搜尋
    National Tsing Hua University Institutional Repository > 電機資訊學院 > 電機工程學系 > 會議論文  >  Video object inpainting using manifold-based posture estimation


    題名: Video object inpainting using manifold-based posture estimation
    作者: Chih-Hung Ling;Yu-Ming Liang;Chia-Wen Lin;Hong-Yuan Mark Liao;Yong-Sheng Chen
    教師: 林嘉文
    日期: 2010
    出版者: Institute of Electrical and Electronics Engineers
    關聯: Proc. IEEE Int. Conf. Image Processing, Hong Kong, China, September 26-29, 2010
    關鍵詞: video inpainting
    object completion
    action prediction
    synthetic posture
    motion animation
    摘要: This paper presents a novel scheme for object completion in a video. The framework includes three steps: posture synthesis, graphical model construction, and action prediction. In the very beginning, a posture synthesis method is adopted to enrich the number of postures. Then, all postures are used to build a graphical model of object action which can provide possible motion tendency. We define two constraints to confine the motion continuity property. With the two constraints, possible candidates between every two consecutive postures are significantly reduced. Finally, we apply the Markov Random Field model to perform global matching. The proposed approach can effectively maintain the temporal continuity of the reconstructed motion. The advantage of this action prediction strategy is that it can handle the cases such as non-periodic motion or complete occlusion.
    相關連結: http:/
    顯示於類別:[電機工程學系] 會議論文
    [光電研究中心] 會議論文


    檔案 描述 大小格式瀏覽次數


    SFX Query


    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - 回饋