Method for processing image with depth information and computer program product thereof
AssigneeNational Chiao Tung University
InventorsJiun-In GUO, Po-Yu CHIEN
Life of this patent
Estimating the term…
Abstract
The present invention proposes a method for processing an image with depth information and a computer program product thereof, wherein a filtering template is used to extract a gesture region and filter the image, and wherein the hue values of the pixels of the current gesture region are used to modify the self-adaptive thresholds of the filtering template, and wherein the size of the filtering template at the next time point is modified according to the depth at the current time point and the depth at the former time point.
Claims24
A method for processing an image with depth information, used to process a first image and a second image of a gesture, and comprising steps:
- acquiring a plurality of feature points of said first image and said second image, and calibrating said first image and said second image to generate a first calibrated image and a second calibrated image;
- determining skin-color blocks and possible skin-color blocks of said first calibrated image and said second calibrated image, distributing a filtering template to said skin-color blocks and said possible skin-color blocks, binarizing said first calibrated image and said second calibrated image to generate a first binarized image and a second binarized image;
- detecting and tracking a first motion part of said first binarized image and a second motion part of said second binarized image, and working out a first positional coordinate set of said first motion part and a second positional coordinate set of said second binarized image;
- comparing said first positional coordinate set and said second positional coordinate set to work out a depth of said gesture;
- and
- comparing one said depth at a current time point and one said depth at a former time point to modify a size of said filtering template at a next time point.
A computer program product storing a program for processing an image with depth information, wherein after said computer program product is loaded into a computer, said computer cam execute a method for processing an image with depth information, used to process a first image and a second image of a gesture, and comprising steps:
- acquiring a plurality of feature points of said first image and said second image, and calibrating said first image and said second image to generate a first calibrated image and a second calibrated image;
- determining skin-color blocks and possible skin-color blocks of said first calibrated image and said second calibrated image, distributing a filtering template to said skin-color blocks and said possible skin-color blocks, binarizing said first calibrated image and said second calibrated image to generate a first binarized image and a second binarized image;
- detecting and tracking a first motion part of said first binarized image and a second motion part of said second binarized image, and working out a first positional coordinate set of said first motion part and a second positional coordinate set of said second binarized image;
- comparing said first positional coordinate set and said second positional coordinate set to work out a depth of said gesture;
- and
- comparing one said depth at a current time point and one said depth at a former time point to modify a size of said filtering template at a next time point.
Dependencies are inferred from the claim text.