Method to detect and assist user intentions with real time visual feedback based on interaction language constraints and pattern recognition of sensory features

• US 8,170,869 B2
• Filed: 10/24/2007
• Issued: 05/01/2012
• Est. Priority Date: 06/28/2007
• Status: Active Grant

• Current Assignee: Sovereign Peak Ventures, LLC (Dominion Harbor Enterprises, LLC)
• Original Assignee: Panasonic Corporation (Panasonic Holdings Corporation)
• Inventors: Rigazio, Luca, Kryze, David, Morin, Philippe
• Primary Examiner(s): Armstrong, Angela A
• Application Number:
  US11/977,331
• Publication Number:
  US 20090006101A1
• Time in Patent Office: 1,651 Days
• Field of Search: 704/9, 704/270, 704/275, 715/700, 715/812, 715/856
• US Class Current: 704/9
• CPC Class Codes: G06F 3/0237 using prediction or retrieval techniques
• CPC Class Codes: G06F 3/04886 by partitioning the display area of the touch-screen or the surface of the digitising tablet into independently controllable areas, e.g. virtual keyboards or menus
• CPC Class Codes: G06F 9/451 Execution arrangements for user interfaces

Abstract

A language model back-off system can be used with a user interface employing one or more language models to constrain navigation of selectable user interface input components. A user input interpretation module receives user input and interprets the user input to determine if a selection is made of one or more user interface input components. If a selection is not made, the user input interpretation module determines whether conditions are met for backing off one or more language models employed to constrain navigation of the user interface input components. If the conditions are met, a language model back-off module backs off the one or more language models.

Claims

1. A language model back-off system for controlling user interaction with a user interface employing plural language models, selected from the group consisting of N-words, N-characters and combinations thereof, comprising:

a navigation constraint module employing plural probabilistic language models, including a first language model and a second language model that provides a lower level of constraint than the first language model, to constrain navigation of selectable user interface input components each corresponding to a corresponding one of characters, so as to constrain the characters that a user is allowed to input,

a user input interpretation module receiving user input and initially interpreting the user input using a level of constraint favoring the first language model to determine if a selection of a next character to be input is made of one or more user interface input components,

wherein said user input interpretation module, if the selection is not made, determines whether conditions are met for backing off the language models employed to constrain navigation of the user interface input components; and

a language model back-off module backing off the language models to change the level of constraint by favoring the second language model over the first language model if the conditions are met, wherein the user input interpretation module and the language model back-off module are implemented by controller.

2. The system of claim 1, further comprising a user interface displaying the user interface input components with a focus indicator.

3. The system of claim 2, further comprising a navigation constraint module employing the plural language models, based on previous selections, to constrain navigation of the selectable user interface input components.

4. The system of claim 3, wherein said navigation constraint module constrains navigation of the user interface input components by controlling sizes of the components based on probabilities of the components being next selected as determined by employing the plural language models.

5. The system of claim 3, wherein said navigation constraint module constrains navigation of the user interface input components by controlling attraction of a pointer to the components based on probabilities of the components being next selected as determined by employing the plural language models.

6. The system of claim 1, wherein said language model back-off module backs off the plural language models by adjusting weights of probabilities that are computed using multiple language models.

7. The system of claim 6, wherein said language model back-off module adjusts the weights to ensure that probabilities computed by employing a more highly constrained language model are given less weight while probabilities computed by employing a less highly constrained language model are given more weight.

8. The system of claim 1, wherein said user input interpretation module determines whether the conditions are met based on a time since a previous selection.

9. The system of claim 1, wherein said user input interpretation module determines whether the conditions are met by employing a probabilistic model taking into account a feature set of sensed user interaction with an input device of the user interface.

10. The system of claim 9, wherein said user input interpretation module employs a feature set including at least one of the following sensory inputs:

(a) a finger position;

(b) a click event;

(c) an acceleration derivative;

(d) a holding integral;

(e) dual finger positions and their multiple derivatives;

(f) dual click events;

(g) dual finger pressure values;

(h) input device acceleration and its integrals;

or

(i) holding sensors outputs and their derivatives.

11. The language model back-off system according to claim 1,

wherein the plural language models are models each employed to provide a probability for selection of a next character, and

the language model back-off module backs off the plural language models if the conditions are met, so that the number of characters provided with a non-zero probability is increased.

12. A language model back-off method for controlling a user interface employing plural language models selected from the group consisting of N-words, N-characters and combinations thereof, comprising:

using a plurality of probabilistic language models including a first language model and a second language model that provides a lower level of constraint than the first language model to constrain navigation of selectable user interface components;

using said first language model to constrain navigation of selectable user interface input components each corresponding to a corresponding one of characters, so as to constrain the characters that a user is allowed to input;

receiving user input and interpreting the user input to determine if a selection of a next character to be input is made of one or more user interface input components;

if the selection is not made, determining whether conditions are met for backing off the language models employed to constrain navigation of the user interface input components; and

if the conditions are met, backing off the plural language models to a lower level of constraint by favoring the second language model over the first language model.

13. The method of claim 12, further comprising displaying the user interface input components with a focus indicator.

14. The method of claim 13, further comprising employing the plural language models, based on previous selections, to constrain navigation of the selectable user interface input components.

15. The method of claim 14, further comprising constraining navigation of the user interface input components by controlling sizes of the components based on probabilities of the components being next selected as determined by employing the plural language models.

16. The method of claim 14, further comprising constraining navigation of the user interface input components by controlling attraction of a pointer to the components based on probabilities of the components being next selected as determined by employing the plural language models.

17. The method of claim 12, further comprising backing off the plural language models by adjusting weights of probabilities that are computed using multiple language models.

18. The method of claim 17, further comprising adjusting the weights to ensure that probabilities computed by employing a more highly constrained language model are given less weight while probabilities computed by employing a less highly constrained language model are given more weight.

19. The method of claim 12, further comprising determining whether the conditions are met based on a time since a previous selection.

20. The method of claim 12, further comprising determining whether the conditions are met by employing a probabilistic model taking into account a feature set of sensed user interaction with an input device of the user interface.

21. The method of claim 20, further comprising employing a feature set including at least one of the following sensory inputs:

(a) a finger position;

(b) a click event;

(c) an acceleration derivative;

(d) a holding integral;

(e) dual finger positions and their multiple derivatives;

(f) dual click events;

(g) dual finger pressure values;

(h) input device acceleration and its integrals;

or

(i) holding sensors outputs and their derivatives.

22. A language model back-off system for use with a user interface employing plural language models, selected from the group consisting of N-words, N-characters and combinations thereof and are employed to constrain navigation of selectable user interface input components each corresponding to a corresponding one of characters, so as to constrain the characters that a user is allowed to input, the language model back-off system comprising:

a user interface displaying the user interface input components with a focus indicator;

a navigation constraint module having plural probabilistic language models including a first language model and a second language model that provides a lower level of constraint that the first language model, and operative based on previous selections to initially constrain navigation of the selectable user interface input components using the first language model;

a user input interpretation module receiving user input and interpreting the user input to determine if a selection of a next character to be input is made of the plural user interface input components, wherein said user input interpretation module, if the selection is not made, determines whether conditions are met for backing off the plural language models based on;

(a) a time since a previous selection; and

(b) a probabilistic model taking into account a feature set of sensed user interaction with an input device of the user interface; and

a language model back-off module backing off the plural probabilistic language models to change the level of constraint by favoring the second language model if the conditions are met and by adjusting weights of probabilities that are computed using the plural probabilistic language models to ensure that probabilities computed by employing a more highly constrained language model are given less weight while probabilities computed by employing a less highly constrained language model are given more weight.