Pointer projection for natural user input
AssigneeMicrosoft Technology Licensing, LLC
InventorsMark Schwesinger, Tommer Leyvand, Szymon Stachniak
Life of this patent
Estimating the term…
Abstract
A method to identify a targeted object based on eye tracking and gesture recognition. The method is enacted in a compute system controlled by a user and operatively coupled to a machine vision system. In this method, the compute system receives, from the machine vision system, video imaging a head and pointer of the user. Based on the video, the compute system computes a geometric line of sight of the user, which is partly occluded by the pointer. Then, with reference to position data for one or more objects, the compute system identifies the targeted object, situated along the geometric line of sight.
Claims20
In a compute system operatively coupled to a machine vision system, a method comprising:
- receiving, from the machine vision system, video imaging a head and pointer of a user in the same field of view;
- computing, based on the video, a geometric line of sight of the user partly occluded by the pointer;
- identifying, with reference to position data for one or more objects, a targeted object situated along the geometric line of sight;
- processing the video to identify an incremental forward movement of the pointer along the geometric line of sight;
- and
- executing a process associated with the targeted object in response to the incremental forward movement.
A compute system comprising a logic machine and a data-storage machine, the data-storage machine storing instructions executable by the logic machine to:
- receive from the machine vision system, via a hardware interface, video imaging a head and pointer of the user;
- compute, based on the video, an initial line of sight of the user partly occluded by the pointer;
- identify, with reference to position data for one or more registered objects, a targeted object situated along the initial line of sight;
- compute, based on the video, a subsequent line of sight of the user partly occluded by the pointer;
- and
- form on the display, along the subsequent line of sight, an image to represent the targeted object.
In a compute system operatively coupled to a machine vision system and to a display, a method comprising:
- receiving, from the machine vision system, video imaging a head and hand of the user;
- computing, based on the video, an initial line of sight of the user partly occluded by the hand;
- identifying, with reference to position data for one or more objects, a targeted object situated along the initial line of sight;
- processing the video to identify movement of the hand away from the user, along the line of sight;
- shifting a depth coordinate of the image in response to movement of the hand away from the user, along the line of sight;
- processing the video to identify closure of the hand, with the hand remaining in the line of sight;
- and
- reducing the size of the image in response to closure of the hand, with the hand maintained in the line of sight.
Dependencies are inferred from the claim text.