REPRESENTING MOVING IMAGES WITH LAYERS

被引：591
作者：
WANG, JYA
ADELSON, EH
机构：
[1] MIT,DEPT BRAIN & COGNIT SCI,CAMBRIDGE,MA 02139
[2] MIT,MEDIA LAB,VIS & MODELING GRP,CAMBRIDGE,MA 02139
来源：
IEEE TRANSACTIONS ON IMAGE PROCESSING | 1994年 / 3卷 / 05期
关键词：
Alpha map - Image sequences - Intensity map - Layered representation - Motion analysis - Overlapping layers - Velocity map;
D O I：
10.1109/83.334981
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
We describe a system for representing moving images with sets of overlapping layers. Each layer contains an intensity map that defines the additive values of each pixel, along with an alpha map that serves as a mask indicating the transparency. The layers are ordered in depth and they occlude each other in accord with the rules of compositing. Velocity maps define how the layers are to be warped over time. The layered representation is more flexible than standard image transforms and can capture many important properties of natural image sequences. We describe some methods for decomposing image sequences into layers using motion analysis, and we discuss how the representation may be used for image coding and other applications.
引用
收藏
页码：625 / 638
页数：14
相关论文
共 23 条
  • [1] ADELSON EH, 1991, MIT181 MED LAB TECH
  • [2] BERGEN JR, 1990, THIRD INTERNATIONAL CONFERENCE ON COMPUTER VISION, P27
  • [3] BERGEN JR, 1992, 2ND P EUR C COMP VIS, P237
  • [4] BLACK MJ, 1991, JUN P COMP VIS PATT, P296
  • [5] DARRELL T, 1991, OCT IEEE WORKSH VIS, P173
  • [6] DEPOMMIER R, 1992, APR IEEE ICASSP SAN, V3, P269
  • [7] DESAI U, 1994, THESIS MIT CAMBRIDGE
  • [8] STOCHASTIC RELAXATION, GIBBS DISTRIBUTIONS, AND THE BAYESIAN RESTORATION OF IMAGES
    GEMAN, S
    GEMAN, D
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1984, 6 (06) : 721 - 741
  • [9] Huang T.S., 1981, IMAGE SEQUENCE ANAL, P289
  • [10] IRANI M, 1992, JUN P IEEE C COMP VI, P216
123