User centric interface for interaction with visual display that recognizes user intentions
First Claim
Patent Images
1. A system for identifying a target object that is displayed on a screen, the system comprising at least one processor that:
- determines a position of a first body part of a user using a first video frame from a camera;
determines a position of a second body part of the user using the first video frame;
identifies a region of a screen based on the position of the first body part and the position of the second body part;
identifies a target object, from among graphical objects displayed on the screen, that corresponds to the region on the screen;
identifies a first point on an image plane associated with the camera, wherein the first point corresponds to a location of the second body part along a calibration sight line between the first body part and a calibration object displayed on the screen;
identifies a second point on the image plane, wherein the second point corresponds to a location of the second body part along a target sight line between the first body part and a target point on the screen;
identifies a third point on a second plane that corresponds to the second point on the image plane, wherein the second plane is parallel to the screen and intersects the image plane;
identifies a fourth point on the second plane that corresponds to the first point on the image plane;
determines a scaling factor based on a position of the calibration object relative to a position of the camera, and further based on a location of the fourth point on the second plane; and
identifies the target object based on the scaling factor and the location of the third point on the second plane.
1 Assignment
0 Petitions
Accused Products
Abstract
Systems, methods, means and computer program products for identifying graphical objects are disclosed. Certain systems, methods, means and computer program products may identify a graphical object based on geometric relationships between two or more user body parts and the graphical object. Certain systems, methods, means and computer program products identify a graphical object based on analysis of pictorial images depicting positions of user body parts relative to each other.
-
Citations
7 Claims
-
1. A system for identifying a target object that is displayed on a screen, the system comprising at least one processor that:
-
determines a position of a first body part of a user using a first video frame from a camera; determines a position of a second body part of the user using the first video frame; identifies a region of a screen based on the position of the first body part and the position of the second body part; identifies a target object, from among graphical objects displayed on the screen, that corresponds to the region on the screen; identifies a first point on an image plane associated with the camera, wherein the first point corresponds to a location of the second body part along a calibration sight line between the first body part and a calibration object displayed on the screen; identifies a second point on the image plane, wherein the second point corresponds to a location of the second body part along a target sight line between the first body part and a target point on the screen; identifies a third point on a second plane that corresponds to the second point on the image plane, wherein the second plane is parallel to the screen and intersects the image plane; identifies a fourth point on the second plane that corresponds to the first point on the image plane; determines a scaling factor based on a position of the calibration object relative to a position of the camera, and further based on a location of the fourth point on the second plane; and identifies the target object based on the scaling factor and the location of the third point on the second plane. - View Dependent Claims (2, 3, 4, 5, 6, 7)
-
Specification