Method and Device for Identifying and Extracting Images of multiple Users, and for Recognizing User Gestures
First Claim
1. A method for identifying and extracting images of one or more users in an interactive environment comprising the steps of:
- obtaining a depth map (7) of a scene in the form of an array of depth values, and an image (8) of said scene in the form of a corresponding array of pixel values, said depth map (7) and said image (8) being registered;
applying a coordinate transformation to said depth map (7) and said image (8) for obtaining a corresponding array (15) containing the 3D positions in a real-world coordinates system and pixel values points;
grouping said points according to their relative positions, by using a clustering process (18) so that each group contains points that are in the same region of space and correspond to a user location (19);
defining individual volumes of interest (20) each corresponding to one of said user locations (19);
selecting, from said array (15) containing the 3D positions and pixel values, the points located in said volumes of interest for obtaining segmentation masks (35) for each user;
applying said segmentation masks (35) to said image (8) for extracting images of said users.
1 Assignment
0 Petitions
Accused Products
Abstract
The invention relates to a method for identifying and extracting images of one or more users in an interactive environment comprising the steps of: —obtaining a depth map (7) of a scene in the form of an array of depth values, and an image (8) of said scene in the form of a corresponding array of pixel values, said depth map (7) and said image (8) being registered; applying a coordinate transformation to said depth map (7) and said image (8) for obtaining a corresponding array (15) containing the 3D positions in a real-world coordinates system and pixel values points; —grouping said points according to their relative positions, by using a clustering process (18) so that each group contains points that are in the same region of space and correspond to a user location (19); —defining individual volumes of interest (20) each corresponding to one of said user locations (19); —selecting, from said array (15) containing the 3D positions and pixel values, the points located in said volumes of interest for obtaining segmentation masks (35) for each user; —applying said segmentation masks (35) to said image (8) for extracting images of said users. The invention also relates to a method for recognizing gestures of said users.
331 Citations
19 Claims
-
1. A method for identifying and extracting images of one or more users in an interactive environment comprising the steps of:
-
obtaining a depth map (7) of a scene in the form of an array of depth values, and an image (8) of said scene in the form of a corresponding array of pixel values, said depth map (7) and said image (8) being registered; applying a coordinate transformation to said depth map (7) and said image (8) for obtaining a corresponding array (15) containing the 3D positions in a real-world coordinates system and pixel values points; grouping said points according to their relative positions, by using a clustering process (18) so that each group contains points that are in the same region of space and correspond to a user location (19); defining individual volumes of interest (20) each corresponding to one of said user locations (19); selecting, from said array (15) containing the 3D positions and pixel values, the points located in said volumes of interest for obtaining segmentation masks (35) for each user; applying said segmentation masks (35) to said image (8) for extracting images of said users. - View Dependent Claims (2)
-
-
3-13. -13. (canceled)
-
14. A device for identifying and extracting images of multiple users in an interactive environment scene comprising:
-
a video camera (4) for capturing an image from the scene; a depth perception devices (3) for providing depth information about said scene; at least one computer processor (2) for processing said depth information and said image information; characterized in that it comprises means (2) for using individual volumes of interest (20) from said scene for each user (6, 6′
).- View Dependent Claims (15)
-
-
16. (canceled)
- 17. Use of a method and/or of a device for identifying and segmenting one or more users in an interactive environment for providing an interactive virtual reality game wherein a plurality of users are separated from each other and from a background.
-
19-28. -28. (canceled)
Specification