System & Method for Constructing, Augmenting & Rendering Multimedia Stories

• US 20150220537A1
• Filed: 02/04/2015
• Published: 08/06/2015
• Est. Priority Date: 02/06/2014
• Status: Active Application

• Current Assignee: Kibra, LLC
• Original Assignee: Kibra, LLC
• Inventors: COSGROVE, Patrick A., SASSON, Steven J., McCOY, John R., GIBSON, Ben K., SCHNEIDER, Mark A., SCHILLE, Paul E., BENNETT, Eiryanna M.K.
• Application Number:
  US14/614,245
• Publication Number:
  US 20150220537A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/438 Presentation of query results

Abstract

Systems to manage the collection of multimedia and physical-sensor data automatically strike a balance between resources utilized to perform the collection, and the usefulness of the data recorded. The recorded data may be augmented with additional data or data streams, and the resulting dataset is automatically composited to produce an audience presentation focusing on selected aspects of the full dataset. The same source dataset may be composited differently to produce another presentation with a different purpose, theme or focus.

Claims

1. A method for building a Event Kernel, comprising:

acquiring a plurality of contemporaneous data streams describing physical conditions measured during an event, an acquisition rate of said acquiring governed by a heuristic taking as inputs at least one data point from at least one of the plurality of contemporaneous data streams and at least one measure of a limiting resource; and

storing the acquired data.

2. The method of claim 1 wherein the plurality of contemporaneous data streams are chosen from the set consisting of {photo, video, audio, GPS, accelerometer, WiFi beacons, temperature, pressure, compass, heart rate, breathing, Date/Time, skin galvanic response, infrared detectors, humidity, control button state, atmospheric pressure vs. blood pressure}.

3. The method of claim 1 wherein at least one data stream of the plurality of contemporaneous data streams is acquired by receiving a wireless transmission from a device that measures a physical condition.

4. The method of claim 3 wherein the wireless transmission is an encrypted wireless transmission.

5. The method of claim 3 wherein the wireless transmission includes a security authentication data exchange.

6. The method of claim 1 wherein a first acquisition rate of a first of the plurality of data streams is different from a second acquisition rate of a second of the plurality of data streams.

7. The method of claim 1, further comprising:

compressing one of the data streams before storing the acquired data.

8. The method of claim 7 wherein a first compression ratio of a first of the plurality of data streams is different from a second compression ratio of a second of the plurality of data streams.

9. The method of claim 7 wherein a compression ratio of the compressed data stream is different at a first time and at a second, later time.

10. The method of claim 1 wherein the heuristic increases the acquisition rate when at least one of the following conditions is detected:

{sound is loud, faces recognized in video, accelerometer peaks, user input received}.

11. The method of claim 1 wherein the heuristic decreases the acquisition rate when at least one of the following conditions is detected:

{sound is quiet, video dark or static, accelerometer idle, battery low, storage full}.

12. The method of claim 1 wherein the heuristic incorporates an indication received from a user of a device performing the method.

13. The method of claim 1 wherein the heuristic reduces the acquisition rate to zero when a privacy signal is detected.

14. The method of claim 13 wherein the privacy signal can be derived from an outside source.

15. The method of claim 13 wherein the privacy signal can be detected or inferred from the sensor capture Streams

16. The methods of claim 13 wherein the privacy signal can come from an external source in combination with detection or inference from the sensor capture streams

17. The method of claim 1, further comprising:

transmitting a data point from one of the plurality of contemporaneous data streams to a remote server;

receiving a reply from the remote server; and

storing a portion of the reply with the acquired data.

18. The method of claim 17 wherein the transmitting and receiving operations are substantially contemporaneous with an acquisition of the data point.

19. The method of claim 17 wherein the transmitting and receiving operations occur after storing the acquired data.

20. The method of claim 1, further comprising:

deleting a subset of the acquired data.

21. The method of claim 1, further comprising:

receiving an external data stream from a device measuring a physical condition other than one measured in the plurality of contemporaneous data streams; and

adding the external data stream to the plurality of contemporaneous data streams before storing the acquired data.

22. The method of claim 1, further comprising:

receiving an external data stream from a device measuring a physical condition duplicative of one measured in the plurality of contemporaneous data streams; and

adding the external data stream to the plurality of contemporaneous data streams before storing the acquired data.

23. A method for uplifting data in a Event Kernel, comprising:

scanning through substantially all of a data stream of a Event Kernel containing a plurality of data streams representing contemporaneous physical measurements of a plurality of different conditions, said scanning occurring in substantially temporal order;

obtaining a first insight about a first portion of the data stream of the Event Kernel during said scanning operation, said first insight based on contents of a subset of the plurality of data streams occurring before a time of the first portion;

recording the first insight in the Event Kernel;

repeating the scanning operation;

obtaining a second insight about a second portion of the data stream of the Event Kernel during said repeated scanning operation, said second insight based on contents of a subset of the plurality of data streams occurring before a time of the second portion and before the time of the first portion; and

recording the second insight in the Event Kernel.

24. The method of claim 23 wherein the data stream is a video stream and the first insight is an identification of a person shown in the video stream.

25. The method of claim 23 wherein the first insight is a function of two independent but contemporaneous data streams of the Event Kernel.

26. The method of claim 23 wherein the second insight is a function of the data stream and the first insight.

27. A method for creating a presentation from a Event Kernel, comprising:

separating data points from a plurality of data streams of a Event Kernel into included and excluded sets;

storing information to identify the included and excluded sets with the Event Kernel; and

reproducing data points from the included set for display to an audience.

28. The method of claim 27 wherein the information to identify the included and excluded sets comprises an editing script.

29. The method of claim 28 wherein the editing script comprises a plurality of editing-action specifiers.

30. The method of claim 29 where an editing-action specifier is chosen from the set consisting of {select video clip, select audio clip, insert transition, insert subtitle}.

31. The method of claim 27 wherein separating comprises:

examining audio data points from an audio data stream and adding contemporaneous data points from the plurality of data streams to the excluded set when the audio data points indicate substantial silence.

32. The method of claim 27 wherein separating comprises:

examining video data points from a video data stream and adding contemporaneous data points from the plurality of data streams to the excluded set when the video data points indicate little activity.

33. The method of claim 27 wherein separating comprises:

examining audio data points from an audio data stream to identify at least one of laughter or applause and adding contemporaneous data points from the plurality of data streams to the included set when the audio data points indicate one of laughter or applause.

34. The method of claim 27 wherein separating comprises:

examining audio data points from an audio data stream to identify a voice and adding contemporaneous data points from the plurality of data streams to the included set when a voice is identified.

35. The method of claim 27 wherein separating comprises:

examining audio data points from an audio data stream to identify a predetermined word and adding contemporaneous data points from the plurality of data streams to the included set when the predetermined word is identified.

36. The method of claim 27 wherein separating comprises:

examining image data points from a video data stream to detect faces and adding contemporaneous data points from the plurality of data streams to the included set when the faces are detected.

37. The method of claim 36, further comprising:

attempting to recognize a face detected in the video data stream; and

if the recognition attempt is successful, adding an identity of the person whose face was recognized to the Event Kernel.

38. The method of claim 27, further comprising:

collecting a playback data stream during the reproducing operation; and

recording the playback data stream with the plurality of data streams.

39. The method of claim 27, further comprising:

repeating the separating operation to produce a second, different pair of included and excluded sets; and

storing the second pair of included and excluded sets with the Event Kernel.

40. The method of claim 27, further comprising:

recording a playback data stream on a computer-readable medium during the reproducing operation.

41. The method of claim 27 wherein reproducing data points comprises:

creating a graphical figure to represent a scalar measure recorded in the Event Kernel; and

compositing the graphical figure with a video output stream.

42. The method of claim 41 wherein the scalar measure is one of a temperature, a heart rate, a velocity, an acceleration, an altitude, a throttle position or a brake position.

43. The method of claim 27, further comprising:

computing an uplift from data in the Event Kernel and data obtained from a source outside the Event Kernel; and

storing the uplift in the Event Kernel, wherein

the separating operation refers to the uplift for determining whether a data point belongs to the included set or the excluded set.

44. The method of claim 27, further comprising:

retrieving supplemental data from a source outside the Event Kernel, said supplemental data identified by reference to data contained in the Event Kernel, wherein

the reproducing operation includes reproducing a portion of the supplemental data.

45. The method of claim 44 wherein the supplemental data is a geographic map, the data contained in the Event Kernel is GPS data, and wherein

reproducing a portion of the supplemental data is animating a track on the geographic map based on the GPS data in the Event Kernel.

46. The method of claim 27, further comprising:

computing an emotional merit function based on the data streams in the Event Kernel, wherein

the separating operation distinguishes between included and excluded data points by comparing a value of the emotional merit function with a predetermined threshold value.

47. The method of claim 46, further comprising:

receiving a sequence of target emotional merit function values, said values corresponding to phases of a story template, and wherein

the separating operation is to select included data sets that satisfy each phase of the story template.

48. The method of claim 27, further comprising:

receiving commands from a human editor during the separating operation, said commands to alter an automatic decision of which data points to include and which data points to exclude; and

altering the information to identify the included and excluded sets before the storing operation.

49. The method of claim 27, further comprising:

altering an order of data points in the included set so that the included data points are reproduced out of temporal order.