Gesture recognition systems
A method and apparatus for performing gesture recognition. In one embodiment of the invention, the method includes the steps of receiving one or more raw frames from one or more cameras, each of the one or more raw frames representing a time sequence of images, determining one or more regions of the one or more received raw frames that comprise highly textured regions, segmenting the one or more determined highly textured regions in accordance textured features thereof to determine one or more segments thereof, determining one or more regions of the one or more received raw frames that comprise other than highly textured regions, and segmenting the one or more determined other than highly textured regions in accordance with color thereof to determine one or more segments thereof. One or more of the segments are then tracked through the one or more raw frames representing he time sequence of images.
1. A method for clustering portions of an image in accordance with a segmentation process, comprising the steps of:
before determining a depth of one or more portions of the image, defining a cluster map comprising pixels associated with at least a portion of an image each cluster being defined by grouping one or more pixels having similar texture;
assigning each pixel in the cluster map a unique numerical identifier corresponding to a spatial coordinate thereof;
defining a two dimensional window;
assigning a lowest cluster number to all of the pixels within the two dimensional window that do not include indications of a changed segment;
comparing a cluster number associated with each pixel with the spatial coordinate thereof;
if it is determined that the cluster number associated with the pixel does not match the spatial coordinate, tracking the cluster number back to a previous pixel whose spatial coordinate actually matches that cluster number, otherwise maintaining the current cluster number associated with the pixel;
writing to the cluster map for the pixel being evaluated the matching spatial coordinate;
comparing neighboring pixels and storing the lowest cluster number of the neighboring pixels;
after comparing neighboring pixels and storing the lowest cluster number of the neighboring pixels, determining a depth associated with each cluster number on the cluster map, and;
defining a depth map in accordance with the determined depth associated with each cluster number on the cluster map.
2. The method of claim 1 , wherein the indications of a changed segment include changes in at least one of, intensity, color, or other defined pixel characteristic.
3. The method of claim 1 , wherein the spatial coordinates are assigned to consecutive pixels.
4. The method of claim 1 , wherein the step of tracking the cluster number back to a previous pixel further comprises the steps of:
linking to a prior pixel associated with the cluster number associated with a current pixel;
determining whether the cluster number of that prior pixel matches the spatial coordinate thereof; and
if the cluster number of that prior pixel does match the spatial coordinate thereof, writing the cluster number to the cluster number of the current pixel, otherwise performing the linking step to another prior pixel.