초록 ▼

An adaptive interface for a programmable system, for predicting a desired user function, based on user history, as well as machine internal status and context. The apparatus receives an input from the user and other data. A predicted input is presented for confirmation by the user, and the predictiv

An adaptive interface for a programmable system, for predicting a desired user function, based on user history, as well as machine internal status and context. The apparatus receives an input from the user and other data. A predicted input is presented for confirmation by the user, and the predictive mechanism is updated based on this feedback. Also provided is a pattern recognition system for a multimedia device, wherein a user input is matched to a video stream on a conceptual basis, allowing inexact programming of a multimedia device. The system analyzes a data stream for correspondence with a data pattern for processing and storage. The data stream is subjected to adaptive pattern recognition to extract features of interest to provide a highly compressed representation which may be efficiently processed to determine correspondence. Applications of the interface and system include a VCR, medical device, vehicle control system, audio device, environmental control system, securities trading terminal, and smart house. The system optionally includes an actuator for effecting the environment of operation, allowing closed-loop feedback operation and automated learning.

대표청구항 ▼

1. A programmable control responsivefor processing a program, configured to respond to ana user input and to a program signal received from a signal source, comprising: a controller, operatingconfigured to operate according to a predetermined program, for receivingto receive the user input and the p

1. A programmable control responsivefor processing a program, configured to respond to ana user input and to a program signal received from a signal source, comprising: a controller, operatingconfigured to operate according to a predetermined program, for receivingto receive the user input and the program signal and producing, to produce a control output, said controller producing a multivalued characterization of the signalto characterize path-dependent user input information as at least one time-dependent motion gesture with respect to at least one context-dependent multivalued parameter, and to characterize path-independent user information;a feedback device configured to present information to the user dependent on said at least one time-dependent motion gesture and said characterized path-independent user information; a memory for storingconfigured to store data relatingrelated to an activity of thea user; and a data processing system for predicting a mostthe controller being further configured to predict at least one probable action of aintended control input with respect to the user input based on said storedthe data relatingrelated to saidthe activity of the user and said characterized signal; and the at least one time-dependent motion gesture, a user feedback data presenting system comprising a display device for presentation of a sequence of programming options to the user, includingwherein the controller is further configured to selectively respond to selectively respond to said at least one most probable action of the user, in a plurality of display images, each display image differing in available programming optionsinput. 2. The programmable control according to claim 1being for performing an action based on user input and an information content of a signal received from a signal source, further comprising: a user controlled direct manipulation-type interface configured to receive the user input device, associated with said display device, having a device output, said device output being the user input;a plant capable of performing a physical action, being responsive to an actuator signal; andsaid controller, being for receiving data from said device output of said input device and a signal received from a signal source, and displayingand to receive the interface user feedback data on said display device, the interfacesaid user feedback data comprising a presentation of a sequence of programming options to the user on said display device, including said most probable actionintended input of the user, in a plurality of display images, each display image differing in available programming options, said sequence of programming options including at least one sequence of options sufficient to define an operable control program, and a presentation of additional programming options if said control program is not operable. 3. The system according to claim 1, being for processing a program comprising program material, in response to a viewer input, comprising: a user input processing system for determining a viewer preference;said controller comprises a program material processing system for characterizing the program material, as said characterized signal, based on its content;a correlator for correlating said characterized content of the program material with said determined viewer preference to produce a correlation index; anda processor, selectively processing the program material based on said correlation index,wherein said viewer preference is an input to said data processing system, andsaid data processing system comprises said correlator. 4. The systemprogrammable control according to claim 31, wherein said program materialsignal is encrypted, further comprising: a decryption system for decryptingconfigured to selectively decrypt the program material if it is selectedsignal to produce unencrypted program material and optionally an associated decryption event;a memory for storing data relating to the occurrence of said decryption event; anda central database for storing data relating to the occurrence of said decryption event in association with data relating to the viewersignal. 5. The systemprogrammable control according to claim 31, wherein: said user input processing systemcontroller monitors a pattern of user activity and predicts a vieweruser preference; and said program material processing system comprises:a processor for preprocessingsaid controller is configured to preprocess the program materialuser input to produce a reduced data flow information signal substantially retaining information relating to an abstract information content of the program materialuser input and selectively eliminating data not relating to said abstract information content of the program materialuser input and for characterizing said reduced data flow information signal based on said abstract information content; anda comparing system for determining if said correlation index is indicative of a probable high correlation between said characterization of said information signal and said viewer preference and causing said stored program material to be processed by said processor based on said determination. 6. The systemprogrammable control according to claim 31, wherein said processor comprises an imagefurther comprising a copy protection system configured to control a use of the program material storage and retrieval systemsignal. 7. The systemprogrammable control according to claim 31, further comprising a memory for storingconfigured to store a characterization of the program materialuser input; and an input for receivingconfigured to receive a feedback signal from the vieweruser indicating a degree of agreement with said correlation index determination, wherein said feedback signal and said stored characterization are used by said user input processing system to predict a new viewer preferenceuser intended input. 8. The systemprogrammable control according to claim 37, wherein said controller further comprises a program materialsignal processing system correlates and characterizes imageconfigured to correlate and characterize information, further comprising: means for storing template data;means for storing the image data;means for generating a plurality of addressable domains from the stored image data, each of the domains representing a different portion of the image information;means for creating, from the stored image data, a plurality of addressable mapped ranges corresponding to different subsets of the stored image data, the creating means including means for executing, for each of the mapped ranges, a procedure upon the one of the subsets of the stored image data which corresponds to the mapped range;means for assigning identifiers to corresponding ones of the mapped ranges, each of the identifiers specifying for the corresponding mapped range an address of the corresponding subset of stored image data;means for selecting, for each of the domains, the one of the mapped ranges which most closely corresponds according to predetermined criteria;means for representing at least a portion of the image information as a set of the identifiers of the selected mapped ranges; andmeans for selecting, from the stored templates, a template which most closely corresponds to the set of identifiers representing the image informationbased on an image analysis algorithm. 9. The systemprogrammable control according to claim 8 wherein said correspondence of a template and the set of identifiers is determined by a processor executing a predetermined program for executing an algorithm selected from at least one of the group consisting of algorithms for selecting a minimum Hausdorff distance between the mapped range and the domain, for selecting the highest cross-correlation of the mapped range with the domain, and for selecting the lowest mean square error of the difference between the mapped range and the domainimage processing algorithm comprises an image transform. 10. The systemprogrammable control according to claim 9 wherein said correspondence of a template and the set of identifiers is determined by selecting, for each domain, the mapped range with the minimum modified Hausdorff distance calculated as D[db,mrb]+D[1−db,I−mrb], where D is a distance calculated between a pair of sets of data each representative of an image, db is a domain, mrb is a mapped range, I−db is the inverse of a domain, and I-mrb is an inverse of a mapped rangeimage transform comprises a transform having an iterative algorithm. 11. The systemprogrammable control according to claim 8, wherein said means for representing further comprises means for: (a) determining a feature of interest of the image data,(b) selecting a mapped range corresponding to the feature of interest,(c) storing the identifiers of the selected mapped range,(d) selecting a further mapped range corresponding to a portion of image data having a predetermined relationship to the feature of interest, and(e) storing the identifiers of the further mapped rangeimage transform comprises a Fourier transform. 12. The systemprogrammable control according to claim 8, wherein said image data comprises data having three associated dimensions obtained by an image processor executing an algorithm selected from the group consisting of an algorithm for synthesizing a three dimensional representation based on a machine based prediction derived from two dimensional image data, an algorithm for synthesizing a three dimensional representation derived from a time series of pixel images, and an algorithm for synthesizing a three dimensional representation based on a image data representing a plurality of parallax views having at least two dimensions, said set of identifiers representing data in each of the three associated dimensions. 13. A method of programmingoperating a device, comprising the steps of: providing an input for user commands and feedback;storing information relating to the user commands and feedback in a memory;providing a data input for external informationidentifying a user;receiving and storing user inputs comprising at least a time-dependent gesture;predictingdetermining a subsequent user commandinput based on the identification of the user, the stored information relating to user commands and feedback, the time-dependent gesture, and a status of the device, and external information from the data input;presenting the predicted user commandfeedback to the user dependent on the time-dependent gesture and the determined user input;accepting feedback from the user relating to the predicted subsequent user command;modifying the predicted subsequent user command based on at least the feedback; andexecutingemploying the modified predicteddetermined user input as a user command input. 14. The method according to claim 13, further comprising the step of identifying a user and storing an user identifier with said stored information relating to the user commands; said predicting step predicting a subsequent user command based on the stored information relating to user identifier, commands and feedback, a status of the device, and external information from the data input. 15. A programmable device comprising: a user input for receivingconfigured to receive an input variable having a path between a first input state and a second input state, said user input comprising: path dependent user data having transitional path information between said first input state and said second input state; andpath independent user data comprising information about at least one of said first state and said second state;a filter configured to separate, separating sadsaid path dependent user data as user characterization data and said path independent user data as instructions;a memory for storingconfigured to store said user characterization data;a processor for executingconfigured to execute said instructions; anda feedback device, presentingconfigured to present information relating to said instructions and said stored user characterization data. 16. The programmable device according to claim 15, further comprising: a hierarchical command structure of said processor, said command structure having commands of different function; andmeans for predicting a probability of execution of a plurality of commands based on said input,said feedback device presenting commands based on at least said predicted probabilities. 17. The programmable control according to claim 115, wherein the signal comprises image informationsaid path dependent user data comprises a gesture. 18. The method according to claim 13, wherein said stored information comprisesA method of operating a device, comprising: receiving and storing information relating to path-dependent gesture user inputs;determining an intended user input based on at least the stored information relating to the path-dependent gesture user inputs, and a status of the device;presenting feedback to the user dependent on the determined intended user input and the stored information relating to the path-dependent gesture inputs;accepting feedback from the user as a basis to modify the determined intended user input; andemploying the modified determined intended user input as a user command input,wherein determining is dependent on a set of weights of a predictive algorithm. 19. The method according to claim 13, herein said external information comprises a perceptual data stream18, further comprising distinguishing between user-intended path-dependent gesture user inputs and user-intended path-independent user inputs. 20. The programmable device according to claim 15, wherein a format of information presented from said feedback device is based on said path dependent user data. 21. A system responsive to a user input and a signal received from a signal source, comprising: a controller configured to operate according to a predetermined program, to receive the input and the signal, to produce a control output, and to produce a multi-valued, content-based characterization of the signal with respect to at least one parameter;a memory configured to store data relating to an activity of a user;a data processing system configured to predict a most probable action of a user based on said stored data relating to said activity of the user and said multi-valued, content-based characterization of the signal; anda user feedback data presenting system configured to present a sequence of programming options to the user, including said most probable action of the user, each option differing in available programming options, wherein said signal comprises image data having three associated dimensions obtained by an image processor executing an algorithm selected from the group consisting of an algorithm for synthesizing a three dimensional representation based on a machine based prediction derived from two dimensional image data, an algorithm for synthesizing a three dimensional representation derived from a time series of pixel images, and an algorithm for synthesizing a three dimensional representation based on a image data representing a plurality of parallax views having at least two dimensions, said set of identifiers representing data in each of the three associated dimensions. 22. The system according to claim 21 being for performing an action based on input and an information content of a signal received from a signal source, further comprising: a user controlled direct manipulation-type input device, associated with a display device, having a device output, said device output being the input; andsaid controller, being for receiving data from said device output of said input device and a signal received from a signal source, and displaying user feedback data on said display device, said user feedback data comprising a presentation of a sequence of programming options to the user on said display device, including said most probable action of the user, in a plurality of display images, each display image differing in available programming options, said sequence of programming options including at least one sequence of options sufficient to define an operable control program, and a presentation of additional programming options if said control program is not operable. 23. The system according to claim 21, being for processing a program comprising content information, comprising: an input processing system for determining a viewer preference;said controller comprises a content information processing system for characterizing the content information, as said characterized signal, based on its content;a correlator for correlating said characterized content of the content information with said determined user preference to produce a correlation index; anda processor, selectively processing the content information based on said correlation index, wherein said user preference is an input to said data processing system, and said data processing system comprises said correlator. 24. The system according to claim 23, wherein said content information is encrypted, further comprising: a decryption system for decrypting the content information if it is selected to produce unencrypted content information and optionally an associated decryption event;a memory for storing data relating to the occurrence of said decryption event; anda central database for storing data relating to the occurrence of said decryption event in association with data relating to the user. 25. The system according to claim 23, wherein: said input processing system monitors a pattern of user activity and predicts a viewer preference for visual content information;said program material processing system comprises:a processor for preprocessing the visual content information to produce a reduced data flow information signal substantially retaining information relating to an abstract information content of the visual content information and selectively eliminating data not relating to said abstract information content of the visual content information and for characterizing said information signal based on said abstract information content; anda comparing system for determining if said correlation index is indicative of a probable high correlation between said characterization of said information signal and said viewer preference and causing said stored program material to be processed by said processor based on said determination. 26. The system according to claim 23, wherein said processor comprises an image program material storage and retrieval system. 27. The system according to claim 23, further comprising: a memory for storing a characterization of the content information;an input for receiving a feedback signal from the user indicating a degree of agreement with said correlation index determination, wherein said feedback signal and said stored characterization are used by said input processing system to predict a new user preference. 28. The system according to claim 23, wherein said content information processing system correlates and characterizes image information, further comprising: a memory adapted to store template data;a memory adapted to store the image data;at least one processor adapted to:generate a plurality of addressable domains from the stored image data, each of the domains representing a different portion of the image information;create, from the stored image data, a plurality of addressable mapped ranges corresponding to different subsets of the stored image data;execute, for each of the mapped ranges, a procedure upon the one of the subsets of the stored image data which corresponds to the mapped range;assign identifiers to corresponding ones of the mapped ranges, each of the identifiers specifying for the corresponding mapped range an address of the corresponding subset of stored image data;select, for each of the domains, the one of the mapped ranges which most closely corresponds according to predetermined criteria;represent at least a portion of the image information as a set of the identifiers of the selected mapped ranges; andselect, from the stored templates, a template which most closely corresponds to the set of identifiers representing the image information. 29. The system according to claim 28 wherein said correspondence of a template and the set of identifiers is determined by at least one processor executing a predetermined program for executing an algorithm selected from at least one of the group consisting of algorithms for selecting a minimum Hausdorff distance between the mapped range and the domain, for selecting the highest cross-correlation of the mapped range with the domain, and for selecting the lowest mean square error of the difference between the mapped range and the domain. 30. The system according to claim 29 wherein said correspondence of a template and the set of identifiers is determined by selecting, for each domain, the mapped range with the minimum modified Hausdorff distance calculated as D[db,mrb]+D[1−db,1−mrb], where D is a distance calculated between a pair of sets of data each representative of an image, db is a domain, mrb is a mapped range, 1−db is the inverse of a domain, and 1−mrb is an inverse of a mapped range. 31. The system according to claim 28, wherein said at least one processor is configured to: determine a feature of interest of the image data,select a mapped range corresponding to the feature of interest,store the identifiers of the selected mapped range,select a further mapped range corresponding to a portion of image data having a predetermined relationship to the feature of interest, andstore the identifiers of the further mapped range. 32. The method according to claim 13, wherein the time-dependent gesture comprises a touch input gesture. 33. The system according to claim 21, wherein the signal comprises image information. 34. A method of interacting with a device, comprising: receiving a user input and feedback from a user relating to an automated output of a predicted input;presenting feedback to the user dependent on at least a time or path dependent gesture represented in the user input and a user instruction represented in the user input;identifying the user;storing a user identifier with information relating to the predicted input, comprising a set of weights of a predictive algorithm and feedback from the user in a memory;storing external information automatically derived from an information content of a signal received from a signal source in a memory;predicting a subsequent input based on the stored information relating to the predicted input and prior feedback, a status of the device, and stored external information;presenting at least one predicted subsequent input to the user;accepting feedback from the user relating to the at least one predicted subsequent input; andstoring the accepted predicted input. 35. The method according to claim 34, further comprising identifying a user and storing a user identifier with said stored information relating to the user commands; said predicting being based on the stored information relating to user identifier, stored information and feedback, a status of the device, and stored external information. 36. The method according to claim 34, wherein the user input comprises a time-dependent gesture. 37. The method according to claim 34, wherein said external information comprises a perceptual data stream. 38. A device comprising: a user input configured to receive an input having a path between a first input state and a second input state, said user input being processed by an automated processor to extract: path dependent user data having transitional path information between said first input state and said second input state; andpath independent user data comprising information about at least one of said first state and said second state;wherein said automated processor is configured to process said path dependent user data for analysis as a user gesture, and to process said path independent user data for analysis as a user instruction; anda feedback device configured to present information to the user dependent on said analysis as a user gesture and said analysis as a user instruction. 39. The device according to claim 38, further comprising: a hierarchical command structure of said at least one processor, said command structure having commands of different function;said at least one processor being adapted to predict a probability of execution of a plurality of commands based on said input, said feedback device presenting commands based on at least said predicted probabilities. 40. The device according to claim 38, wherein a format of information presented from said feedback device is based on said path dependent user data. 41. A control method, comprising: automatically producing a multi-valued characterization of an information content of a received signal;identifying a user and storing a user identifier with data relating to received user inputs;predicting at least one probable action based on the stored data comprising the user identifier and the data relating to the received user inputs, and the multi-valued characterization of the information content of the signal; andpresenting of a plurality of differing options, including the predicted at least one probable action, wherein the signal comprises image data processed to have three associated physical dimensions obtained by an algorithm selected from the group consisting of an algorithm for synthesizing a three dimensional representation based on a machine based prediction derived from two dimensional image data, an algorithm for synthesizing a three dimensional representation derived from a time series of pixel images, and an algorithm for synthesizing a three dimensional representation based on a image data representing a plurality of parallax views having at least two dimensions, said set of identifiers representing data in each of the three associated dimensions. 42. The control method according to claim 41, wherein the set of differing options is presented through a graphic user interface and at least a portion of a graphic user interface input is received through a direct manipulation-type input device. 43. The control method according to claim 41, wherein the received input is provided by a user, further comprising processing the received input to determine a user preference. 44. The control method according to claim 41, wherein said predicting comprises correlating the multi-valued characterization of the information content of the signal with the stored data. 45. The control method according to claim 41, wherein the signal comprises media content information, further comprising processing the input to determine a media content preference. 46. The control method according to claim 45, wherein the multi-valued characterization of the information content of the signal comprises a model of the media content information based in its information content. 47. The control method according to claim 45, wherein the media content information is encrypted, further comprising decrypting the media content information. 48. The control method to claim 41, further comprising monitoring a pattern of user activity represented in the received input and predicting a user preference in dependence thereon. 49. The control method to claim 41, further comprising substantially retaining an abstract information content, and selectively eliminating data not relating to said abstract information content, and characterizing the signal based on the retained abstract information content. 50. The control method according to claim 41, wherein the signal comprises image program material. 51. The control method according to claim 41, further comprising receiving feedback from a user indicating a degree of agreement with a predicted at least one probable action wherein the feedback is used to predict a new at least one probable action. 52. The programmable control according to claim 1, wherein the path-dependent user input information comprises touch information. 53. A method of using a device, comprising: receiving a user input comprising a time and path-dependent gesture and a user-defined state input, and a user-specific context for interacting with the device;presenting feedback to the user representing a prediction of a future user input based on at least the received time and path-dependent gesture and the user-define state input, and the user-specific context;receiving feedback from the user relating to the predicted future user input;modifying the predicted future user input based on at least the feedback and persistently storing a user profile over a range of user contexts; andemploying the modified predicted future user input as a user input for interacting with the device. 54. The method according to claim 53, further comprising identifying a user and separately storing a user profile based on past user input and feedback for each of a plurality of different users. 55. The method according to claim 53, further comprising predicting the future user input further in dependence on a device status. 56. The method according to claim 53, wherein the prediction of a future user input is further based on a perceptual data stream. 57. A user interface method, comprising: receiving an input, the input providing path-dependent gesture information representing a user intended command;determining a context;storing a user profile derived from the input for a respective user over a range of contexts;predicting a context-dependent user intended command of the respective user;generating feedback for the user dependent on at least the path-dependent gesture information and the user-intended command; andexecuting the predicted user intended command. 58. The method according to claim 57, wherein the predicted context-dependent user intended command is modified based on user feedback. 59. A user interface method, comprising: receiving a series of user gesture inputs;automatically determining an associated path-dependent information content of respective user gesture inputs representing at least one command;predicting at least one subsequent member of the series of user gesture inputs based on a stored set of previously received user inputs and respective associated path-dependent information content, an information base, and a set of predictive rules;presenting feedback to the user dependent on at least the respective associated path-dependent information content and the at least one represented command of the user gesture inputs;outputting at least one predicted subsequent member of the sequence; andreceiving path-independent information content from the user representing an intention to execute the at least one predicted subsequent member of the sequence as a command. 60. The user interface method according to claim 59, wherein the prediction is subjectively biased in dependence on a set of past user inputs. 61. A method of programming a device, comprising the steps of: providing a user interface configured to receive path-dependent user gesture input and path-independent selections from the user, and to present automated proposals to the user;storing information relating to the path-dependent user gesture input and path-independent selections in a memory;providing a database of stored information;iteratively: receiving a path-dependent user gesture input; andpresenting through the user interface, feedback to the user comprising at least one automatically generated proposal representing an available option, the at least one automatically generated proposal being dependent on at least prior path-dependent user gesture input, path-independent selections from the user, the database, and an automatically determined environmental status;accepting at least one automatically generated proposal by the user through the user interface based on a path-independent selection; andstoring a completed user session input defined based on said iteratively receiving, presenting, and accepting.