| 摘要 |
The invention relates to a method and device for generating a large static image M(n), such as a sprite or a mosaic, from a video sequence including successive video objects. This method comprises a first step for estimating motion parameters related to the current video object V0(n) of the sequence with respect to the previously generated static image M(n-1), a second step for warping this video object on the basis of the estimated motion parameters, a third step for blending the warped video object thus obtained with the previously generated static image M(n-1), and a fourth step for computing, for each picture element of the current video object, a weighting coefficient WWF(n)[x,y] correlated to the error between the warped video object and the static image M(n-1), the blending formula then taking into account said weighting coefficients. According to the invention, the pixel-based weighting map is converted into a region-based one by means of a spatial filtering step based on a morphological segmentation, which allows to select regions that belong to foreground objects and to discard them as outliers before the blending step.
|
