Distance-based spatial representation and prediction systems, methods and computer program products for associative memories

• US 7,574,416 B2
• Filed: 01/14/2005
• Issued: 08/11/2009
• Est. Priority Date: 01/16/2004
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Saffron Technology, Inc. (Intel Corporation)
• Inventors: Aparicio, Manuel IV, Cabana, David R.
• Primary Examiner(s): Vincent; David R
• Assistant Examiner(s): BUSS, BENJAMIN J
• Application Number:
  US11/036,641
• Publication Number:
  US 20050163347A1
• Time in Patent Office: 1,670 Days
• Field of Search: 382/103, 706/10, 706/48, 706/58, 706/54, 706/18, 348/142, 348/143, 348/169
• US Class Current: 706/58
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments

Abstract

A location of a missing object is predicted based on past sightings of objects including the missing object, and a new sighting of the objects except for the missing object. For a respective given object in the objects, the past sightings are memorized based on respective distances of respective remaining objects from the respective given object. Distance-based memorization may take place using an agent or associative memory for a respective given object. Then, for a respective given object, except for the missing object, a distance of the missing object from the respective given object is predicted, based on the past sightings that have been memorized and the new sighting, to obtain candidate locations for the missing object. The candidate locations are then disambiguated, to predict the location of the missing object.

Claims

1. A method of predicting a location of a missing object based on a plurality of past sightings of a plurality of objects including the missing object and a new sighting of the plurality of objects except for the missing object, comprising:

associating an agent with each respective given object in the plurality of objects, wherein each agent has an associative matrix memory, wherein the agents interact and cohere as a network;

for each respective given object of at least two of the plurality of objects, the respective agent memorizing the plurality of past sightings based on respective distances of respective remaining objects from each respective given object;

for each respective given object of at least two objects of the plurality of objects except for the missing object, predicting a distance of the missing object from the respective given object based on the past sightings that have been memorized and the new sighting, to obtain at least two candidate locations for the missing object, the at least two candidate locations corresponding to the at least two objects of the plurality of objects except for the missing object;

processing the at least two candidate locations corresponding to the at least two of the plurality of objects except for the missing object to predict the location of the missing object, wherein the respective agent associated with the missing object processes the at least two candidate locations to predict the most likely location for its object; and

presenting, to a user, at least one of the candidate locations, one of the plurality of past sightings, a new sighting of the plurality of objects, and/or the predicted location of the missing object.

2. The method according to claim 1 wherein memorizing the plurality of past sightings based on respective distances of respective remaining objects from each respective given object comprises memorizing a plurality of pairs of the remaining objects that have co-occurred in the plurality of past sightings from a perspective of each respective given object of at least two of the plurality of objects, and respective distances of the respective plurality of pairs of objects from each respective given object of at least two of the plurality of objects.

3. The method according to claim 2 wherein predicting a distance of the missing object from each respective object of at least two of the plurality of objects based on the past sightings that have been memorized and the new sighting, to obtain a plurality of candidate locations for the missing object comprises predicting a distance of the missing object from each respective object of at least two of the plurality of objects based on the plurality of pairs of the remaining objects that have co-Occurred in the plurality of past sightings from the perspective of each respective object of at least two of the plurality of objects that have been memorized, and the respective distances of the respective plurality of pairs of objects from each respective object of at least two of the plurality of objects that have been memorized

4. The method according to claim. 1 wherein memorizing the plurality of past sightings based on respective distances of respective remaining objects from each respective given object is performed by a plurality of associated memories corresponding to each respective given object of at least two of the plurality of objects.

5. The method according to claim 4 wherein predicting a distance of the missing object from each respective object of at least two of the plurality of objects based on the past sightings that have been memorized and the new sighting also is performed by the plurality of associated memories corresponding to each respective given object of at least two of the plurality of objects.

6. The method according to claim 5 wherein processing the at least two corresponding to the at least two of the plurality of objects except for the missing object comprises:

predicting a location of the missing object by the respective associative memory for the missing object, based on the past sightings that have been memorized by the respective associative memory for the missing object and the new sighting; and

comparing the predicted location of the missing object with the at least two candidate locations corresponding to the at least two of the plurality of objects except for the missing object.

7. A computer program product comprising a computer usable storage medium having computer-readable program code embodied in the medium, the computer-readable computer code executing to cause a computer to perform the method of claim 1.

8. A system means for performing the method of claim 1, wherein the system further performs said method.

9. A system of predicting a location of a missing object based on a plurality of past sightings of a plurality of objects including the missing object and a new sighting of the plurality of objects except for the missing object, comprising:

a plurality of associative memories which interact and cohere as a network, each respective one of at least two of the plurality of associative memories is configured to memorize the plurality of past sightings from a perspective of a respective given object of at least two of the plurality of objects, based on respective distances of respective remaining objects from the respective given object, each respective one of the at least two associative memories also being configured to predict a distance of the missing object from the respective given object based on the past sightings that have been memorized and the new sighting, and thereby obtain at least two candidate locations for the missing object, the at least two candidate locations corresponding to at least two associative memories of the plurality of associative memories;

means for processing the at least two candidate locations corresponding to the at least two associative memories to predict the location of the missing object, wherein the respective associative memory associated with the missing object processes the at least two candidate locations to predict the most likely location for its object; and

means for presenting, to a user, the predicted location of the missing object.

10. The system according to claim 9 wherein each respective one of at least two of the plurality of associative memories is configured to memorize the plurality of past sightings based on respective distances of respective remaining objects from the respective given object by memorizing a plurality of pairs of the remaining objects that have co-occurred in the plurality of past sightings-from the perspective of the respective given object, and respective distances of the respective plurality of pairs of objects from the respective given object.

11. The system according to claim 10 wherein each respective one of at least two of the plurality of associative memories is configured to predict a distance of the missing object from the respective given object based on the past sightings that have been memorized and the new sighting, by predicting a distance of the missing object from the respective given object based on the plurality of pairs of the remaining objects that have co-occurred in the plurality of past sightings from the perspective of the respective given objects that have been memorized, and the respective distances of the respective plurality of pairs of objects from the respective given objects that have been memorized.

12. The system according to claim 9 wherein the means for processing the at least two candidate locations corresponding to the at least two associative memories to predict the location of the missing object comprises:

means for controlling the associative memory for the missing object to predict a location of the missing object, based on the past sightings that have been memorized by the associative memory for the missing object and the new sighting; and

means for comparing the predicted location of the missing object with the at least two candidate locations corresponding to the at least two associative memories.

13. The system according to claim 11 wherein the means for processing the at least two candidate locations corresponding to the at least two associative memories to predict the location of the missing object comprises:

means for controlling the associative memory for the missing object to predict a location of the missing object, based on the past sightings that have been memorized by the associative memory for the missing object and the new sighting; and

means for comparing the predicted location of the missing object with the at least two candidate locations corresponding to the at least two associative memories.

14. A computer program product comprising a computer usable storage medium having computer-readable program code embodied in the medium, the computer-readable computer code executing to cause a computer to operate as the system of claim 9.

15. A method of predicting a location of a missing object based on a plurality of past sightings of a plurality of objects including the missing object and a new sighting of the plurality of objects except for the missing object, comprising:

configuring each respective one of at least two of a plurality of associative memories to memorize the plurality of past sightings from a perspective of each respective object of at least two of the plurality of objects, based on respective distances of respective remaining objects from each respective given object, wherein the plurality of associative memories interact and cohere as a network;

configuring each respective one of the at least two associative memories to predict a distance of the missing object from each respective given object based on the past sightings that have been memorized and the new sighting, and thereby obtain at least two candidate locations for the missing object, the at least two candidate locations corresponding to the at least two associative memories;

processing the at least two candidate locations corresponding to the at least two associative memories to predict the location of the missing object, wherein the respective associative memory associated with the missing object processes the at least two candidate locations to predict the most likely location for its object; and

presenting, to a user, at least one of the candidate locations, one of the plurality of past sightings, the new sighting of the plurality of objects, and/or the predicted location of the missing object.

16. The method according to claim 15 wherein configuring each respective one of at least two of the plurality of associative memories to memorize the plurality of past sightings from a perspective of each respective object of at least two of the plurality of objects, based on respective distances of respective remaining objects from each respective given object comprises configuring each respective one of at least two of the plurality of associative memories to memorize the plurality of past sightings based on respective distances of respective remaining objects from each respective object of at least two of the plurality of objects by memorizing a plurality of pairs of the remaining objects that have co-occurred in the plurality of past sightings from the perspective of each respective object of at least two of the plurality of objects, and respective distances of the respective plurality of pairs of objects from each respective object of at least two of the plurality of objects.

17. The method according to claim 16 wherein configuring each respective one of the at least two associative memories to predict a distance of the missing object from each respective object of at least two of the plurality of objects based on the past sightings that have been memorized and the new sighting, and thereby obtain at least two candidate locations for the missing object, the at least two candidate locations corresponding to the at least two associative memories comprises configuring each respective one of the at least two associative memories to predict a distance of the missing object from each respective object of at least two of the plurality of objects based on the plurality of pairs of the remaining objects that have co-occurred in the plurality of past sightings from the perspective of each respective object of at least two of the plurality of objects that have been memorized, and the respective distances of the respective plurality of pairs of objects from each respective object of at least two of the plurality of objects that have been memorized.

18. The method according to claim 15 wherein processing the at least two candidate locations corresponding to the at least two associative memories to predict the location of the missing object comprises:

controlling the associative memory for the missing object to predict a location of the missing object, based on the past sightings that have been memorized by the associative memory for the missing object and the new sighting; and

comparing the predicted location of the missing object with the at least two candidate locations.

19. The method according to claim 17 wherein processing the at least two candidate location corresponding to the at least two associative memories to predict the location of the missing object comprises:

controlling the associative memory for the missing object to predict a location of the missing object, based on the past sightings that have been memorized by the associative memory for the missing object and the new sighting; and

comparing the predicted location of the missing object with the at least two candidate locations.

20. A computer program product comprising a computer usable storage medium having computer-readable program code embodied in the medium, the computer-readable computer code executing to cause a computer to perform the method of claim 15.