Recognizing interactive media input
First Claim
Patent Images
1. A method comprising:
- calibrating an interactive media device to identify and visually track;
an object moved through space by a user;
recording a trajectory of a movement of the object through space extracted from a multi-frame video sequence, the trajectory directly representing a desired handwritten character;
displaying the trajectory on a display device;
based at least in part on a shape of the trajectory, translating the trajectory into two or more suggested characters;
ranking each of the two or more suggested characters based on a likelihood that each of the two or more suggested characters are the desired handwritten character;
simultaneously displaying the two or more suggested characters on the display device so that display positions of each of the two or more suggested characters relative to an endpoint of the trajectory are based on the ranking of each of the two or more suggested characters, wherein the ranking is based on the likelihood that each of the two or more suggested characters are the desired handwritten character; and
receiving selection information indicating which of the two or more suggested characters represents the desired handwritten character.
2 Assignments
0 Petitions
Accused Products
Abstract
Techniques and systems for inputting data to interactive media devices are disclosed herein. In some aspects, a sensing device senses an object as it moves in a trajectory indicative of a desired input to an interactive media device. Recognition software may be used to translate the trajectory into various suggested characters or navigational commands. The suggested characters may be ranked based on a likelihood of being an intended input. The suggested characters may be displayed on a user interface at least in part based on the rank and made available for selection as the intended input.
-
Citations
17 Claims
-
1. A method comprising:
-
calibrating an interactive media device to identify and visually track; an object moved through space by a user; recording a trajectory of a movement of the object through space extracted from a multi-frame video sequence, the trajectory directly representing a desired handwritten character; displaying the trajectory on a display device; based at least in part on a shape of the trajectory, translating the trajectory into two or more suggested characters; ranking each of the two or more suggested characters based on a likelihood that each of the two or more suggested characters are the desired handwritten character; simultaneously displaying the two or more suggested characters on the display device so that display positions of each of the two or more suggested characters relative to an endpoint of the trajectory are based on the ranking of each of the two or more suggested characters, wherein the ranking is based on the likelihood that each of the two or more suggested characters are the desired handwritten character; and receiving selection information indicating which of the two or more suggested characters represents the desired handwritten character. - View Dependent Claims (2, 3, 4, 5, 6, 7)
-
-
8. One or more computer-readable memory devices comprising processor-executable instructions that, when executed, direct a device to perform actions comprising:
-
receiving, from a visual sensing device, a video stream of a trajectory of an object moved through space by a user; receiving an input mode directive to designate object tracking as a handwriting mode; receiving a start input signal to start recording the trajectory of the object through space; receiving a stop input signal to stop the recording of the trajectory while continuing to receive the input signal of the trajectory of the object through space from the visual sensing device or the inertial sensing device, wherein a shape of the recorded trajectory of the object through space between the start input signal and the stop input signal represents a handwritten character; based at least in part on a shape of the recorded trajectory, translating the recorded trajectory into one or more suggested characters; and displaying the one or more suggested characters on a display device. - View Dependent Claims (9, 10, 11, 12, 13, 14)
-
-
15. A method comprising:
-
recording a trajectory of a movement of an object moved through space by a user, wherein the trajectory is extracted from a multi-frame video sequence via a camera, and wherein the trajectory directly represents a desired handwritten character; displaying the trajectory on a display device; based at least in part on a shape of the trajectory, translating the trajectory into a plurality of suggested inputs comprising characters; simultaneously displaying the plurality of suggested inputs on a display device so that display positions of the plurality of suggested inputs relative to an endpoint of the trajectory are based on a likelihood of each of the plurality of suggested inputs being the desired handwritten character; receiving a selection of the desired handwritten character, the desired handwritten character being one of the plurality of suggested inputs; and controlling an interactive media device via the desired handwritten character. - View Dependent Claims (16, 17)
-
Specification