• DocumentCode
    2736257
  • Title

    Fast object detection and segmentation in MPEG compressed domain

  • Author

    Sukmarg, Orachat ; Rao, K.R.

  • Author_Institution
    Dept. of Electr. Eng., Texas Univ., Arlington, TX, USA
  • Volume
    3
  • fYear
    2000
  • fDate
    2000
  • Firstpage
    364
  • Abstract
    We present a fast algorithm for object detection and segmentation in the MPEG compressed domain using color clustering, region merging based on spatiotemporal similarities, background/foreground classification, and pixel edge extraction. The features extracted from the blocks of the segmented object in the compressed domain can be used for fast object tracking and indexing at low level. Moreover, these blocks can be decompressed to obtain details of a specific object in the pixel domain and can be used for high level indexing. By using the proposed algorithm, we can reduce the amount of the information needed to be processed, and therefore, save the computational time, and increase the processing speed. Also we need to perform an inverse DCT on only some parts of the image
  • Keywords
    code standards; data compression; decoding; feature extraction; image classification; image colour analysis; image segmentation; image sequences; object detection; pattern clustering; telecommunication standards; tracking; video coding; MPEG compressed domain; background/foreground classification; color clustering; computational time savings; fast algorithm; fast object indexing; fast object tracking; feature extraction; high level indexing; image decompression; inverse DCT; object detection; object segmentation; pixel domain; pixel edge extraction; processing speed; region merging; spatiotemporal similarities; video sequences; Clustering algorithms; Decoding; Feature extraction; Image segmentation; Indexing; Object detection; Partitioning algorithms; Spatiotemporal phenomena; Transform coding; Video compression;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    TENCON 2000. Proceedings
  • Conference_Location
    Kuala Lumpur
  • Print_ISBN
    0-7803-6355-8
  • Type

    conf

  • DOI
    10.1109/TENCON.2000.892290
  • Filename
    892290