Method of sampling colors of images of a video sequence, and application to color clustering
The method comprises the steps of, successively for each image following a preceding image based on a map of motion vectors that corresponds to the motion from said preceding image toward said following image, building a pixel mask for said following image, applying said pixel mask to the corresponding following image in order to obtain a corresponding masked image that samples the pixels of said following image. The application of this method to color clustering allows the iterative update of the colors clusters with limited computer resources.
1. A method of clustering colors of images of a video sequence, comprising:
clustering colors of a first image of the video sequence to generate a set of color clusters and to distribute said colors in color clusters of said generated set, and,
successively for each image following a preceding image of this sequence,
building a pixel mask that samples pixels of said following image that are associated with a color difference above a threshold, wherein said color difference is computed for each pixel of this pixel mask between a color value of the same pixel in said following image and a color value of the same pixel in a compensated image generated by applying a map of motion vectors to said preceding image, wherein said map of motion vectors corresponds to a motion from said preceding image toward said following image,
applying said pixel mask to said following image in order to obtain a masked image, and
distributing colors of said masked image in color clusters of said generated set, resulting in an updated of the distribution of colors in said color clusters.
2. A color clustering device for clustering colors of images of a video sequence comprising a central processing unit having a processor and a memory, the processor and memory being configured to perform as:
an initial color clustering module that clusters colors of a first image of the video sequence to generate a set of color clusters and to distribute said colors in color clusters of said generated set,
a pixel map builder that builds a pixel mask that samples pixels of an image of said sequence following a preceding image that are associated with a color difference above a threshold, wherein said color difference is computed for each pixel of said pixel mask between a color value of the same pixel in said following image and a color value of the same pixel in a compensated image generated by applying a map of motion vectors to said preceding image, said map of motion vectors corresponding to the motion from said preceding image toward said following image, and
an updating color clustering that applies the pixel mask built by said pixel map builder to said following image in order to obtain a masked image, and that distributes the colors of said masked image in color clusters of said generated set, resulting in an updated of the distribution of colors in said color clusters.
3. A method of clustering colors of images of a video sequence, comprising:
clustering colors of a first image of the video sequence to generate a set of color clusters and to distribute said colors in color clusters of said generated set, and,
successively for each image following a preceding image of this sequence,
building a pixel mask that samples pixels of said following image that are associated with a motion difference above a threshold, wherein said motion vector difference is computed for each pixel of this pixel mask, between a motion vector of the same pixel in said following image computed with reference to a corresponding pixel in said preceding image and motion vectors of neighboring pixels located in the neighborhood of said pixel computed in reference to their corresponding pixels in said preceding image,
applying said pixel mask to said following image in order to obtain a masked image, and
distributing colors of said masked image in color clusters of said generated set, resulting in an updated of the distribution of colors in said color clusters.
4. A color clustering device for clustering colors of images of a video sequence comprising:
a central processing unit having a processor and an associated memory, the processor and memory being configured to perform as:
an initial color clustering module that clusters colors of a first image of the video sequence to generate a set of color clusters and to distribute said colors in color clusters of said generated set,
a pixel map builder that builds a pixel mask that samples pixels of an image of said sequence following a preceding image that are associated with a motion difference above a threshold, wherein said motion vector difference is computed for each pixel of this pixel mask, between a motion vector of the same pixel in said following image computed in reference to a corresponding pixel in said preceding image and motion vectors of neighboring pixels located in the neighborhood of said pixel computed in reference to their corresponding pixels in said preceding image, and
an updating color clustering that applies the pixel mask built by said pixel map builder to said following image in order to obtain a masked image, and that distributes the colors of said masked image in color clusters of said generated set, resulting in an update of the distribution of colors in said color clusters.