Key frames extraction in athletic video

被引：2
作者：
Caccia, G [1 ]
Lancini, R [1 ]
Russo, S [1 ]
机构：
[1] CEFRIEL, Politecn Milano, Milan, Italy
来源：
VISUAL COMMUNICATIONS AND IMAGE PROCESSING 2003, PTS 1-3 | 2003年 / 5150卷
关键词：
key frame detection; indexing; MPEG7;
D O I：
10.1117/12.503095
中图分类号：
O43 [光学];
学科分类号：
070207 ; 0803 ;
摘要：
In this paper, we present an effective framework for features extraction from an athletic sport sequence. We analyze both forward and backward motion vectors from MPEG 2 video sequences for camera movements detection. Features like the beginning and the end of the race and the type of competition are strictly connected to the camera motion. Our algorithm is able to extract the frame number of the investigated feature with very high accuracy.
引用
收藏
页码：2097 / 2104
页数：8
相关论文
共 11 条
  • [1] ABOUGHAZALEH NH, 2000, P VCIP 2000 JUN, P986
  • [2] ARDIZZONE, P IEEE ICIP 1996, P831
  • [3] BOUTHEMY P, CIRCUITS SYSTEM VIDE, V9, P1030
  • [4] DURIK M, 2001, P INT WORKSH CONT BA, P57
  • [5] Key frame selection from MPEG video data
    Gerek, OM
    Altunbasak, Y
    [J]. VISUAL COMMUNICATIONS AND IMAGE PROCESSING '97, PTS 1-2, 1997, 3024 : 920 - 925
  • [6] *ISO IEC, 1996, 13818 ISOIEC
  • [7] Video motion representation for improved content access
    Jeannin, S
    Mory, B
    [J]. IEEE TRANSACTIONS ON CONSUMER ELECTRONICS, 2000, 46 (03) : 645 - 655
  • [8] ODOBETZ JM, 1994, ESTIMATION DETECTION
  • [9] TAN YP, P IEEE ICIP 1995, P406
  • [10] TAN YP, CIRCUITS SYSTEM VIDE, V10, P133
12