Methods, apparatus and articles of manufacture to detect shapes
View Patent ↗Methods, apparatus, and articles of manufacture to detect shapes are disclosed. Example methods disclosed herein include determining a first likelihood of detection for a reference shape at a first distance from an image capturing device, and storing the reference shape in a first group of a plurality of groups in the database based on a comparison of the first likelihood of detection with a first threshold, the first group associated with the first distance. Disclosed examples also include selecting reference shapes of the first group to compare to an object outline detected in an image in response to a query, the query including a depth value associated with the detected object outline, the selecting of the reference shapes of the first group based on the depth value.
1. An apparatus to manage a database for object recognition, the apparatus comprising:
a detector to determine a first likelihood of detection for a reference shape at a first distance from an image capturing device;
a grouper to store the reference shape in a first group of a plurality of groups of reference shapes in the database based on a comparison of the first likelihood of detection with a first threshold, the first group associated with the first distance;
a calculator to aggregate a plurality of depth values associated with an object outline in an image to determine a combined depth value associated with the detected object outline; and
a comparator to limit comparison of the object outline in the image to reference shapes of the first group responsive to the combined depth value corresponding to the first distance.
2. The apparatus of claim 1 , wherein the detector is to determine a second likelihood of detection for the reference shape at a second distance from the image capturing device different than the first distance, and the grouper is to store the reference shape in a second group of the plurality of groups when the second likelihood of detection is greater than a second threshold, the second group associated with the second distance.
3. The apparatus of claim 1 , wherein respective ones of the plurality of groups in the database are associated with respective different distances from the image capturing device, and the grouper is to store the reference shape in ones of the plurality of groups in the database based on comparisons of respective likelihoods of detection at the respective distances with respective thresholds associated with the respective distances, a first one of the respective thresholds different than a second one of the respective thresholds.
4. The apparatus of claim 1 , wherein the grouper is to determine a type of the reference shape, and the first group is also associated with the type of the reference shape.
5. The apparatus of claim 4 , wherein the type of the reference shape includes at least one of a standing person, a head, a chair, and a combination of a person and a couch.
6. The apparatus of claim 1 , wherein the detector is to determine the first likelihood of detection based on at least one of a size of the reference shape or input from an administrator.
7. The apparatus of claim 1 , wherein the detector is to determine the first likelihood of detection based on a Bayesian network.
8. A tangible machine readable storage medium comprising instructions that, when executed, cause a processor to at least:
determine a first likelihood of detection for a reference shape at a first distance from an image capturing device;
store the reference shape in a first group of a plurality of groups of reference shapes in a database based on a comparison of the first likelihood of detection with a first threshold, the first group associated with the first distance;
aggregate a plurality of depth values associated with an object outline in an image to determine a combined depth value associated with the detected object outline; and
limit comparison of the object outline in the image to reference shapes of the first group responsive to the combined depth value corresponding to the first distance.
9. The storage medium of claim 8 , where in the instruction further cause the processor to determine a second likelihood of detection for the reference shape at a second distance from the image capturing device different than the first distance, and store the reference shape in a second group of the plurality of groups when the second likelihood of detection is greater than a second threshold, the second group associated with the second distance.
10. The storage medium of claim 8 , wherein respective ones of the plurality of groups in the database are associated with respective different distances from the image capturing device, and the instructions further cause the processor to store the reference shape in ones of the plurality of groups in the database based on comparisons of respective likelihoods of detection at the respective distances with respective thresholds associated with respective distances, a first one of the respective thresholds different than a second one of the respective thresholds.
11. The storage medium of claim 8 , wherein the instructions further cause the processor to determine a type of the reference shape, and the first group is also associated with the type of the reference shape.
12. The storage medium of claim 11 , wherein the type of the reference shape includes at least one of a standing person, a head, a chair, and a combination of a person and a couch.
13. The storage medium of claim 8 , wherein the instructions further cause the processor to determine the first likelihood of detection based on at least one of a size of the reference shape or input from an administrator.
14. The storage medium of claim 8 , wherein the instructions further cause the processor to determine the first likelihood of detection based on a Bayesian network.
15. A method comprising:
determining, by executing an instruction with a processor, a first likelihood of detection for a reference shape at a first distance from an image capturing device;
storing the reference shape in a first group of a plurality of groups of reference shapes in a database based on a comparison of the first likelihood of detection with a first threshold, the first group associated with the first distance;
aggregating, by executing an instruction with the processor, a plurality of depth values associated with an object outline in an image to determine a combined depth value associated with the detected object outline; and
limiting comparison of the object outline in the image to reference shapes of the first group responsive to the combined depth value corresponding to the first distance.
16. The method of claim 15 , further including determining a second likelihood of detection for the reference shape at a second distance from the image capturing device different than the first distance, and store the reference shape in a second group of the plurality of groups when the second likelihood of detection is greater than a second threshold, the second group associated with the second distance.
17. The method of claim 15 , further including determining a type of the reference shape and associating the first group with the type of the reference shape.
18. The method of claim 17 , wherein the type of the reference shapes includes at least one of a standing person, a head, a chair, and a combination of a person and a couch.
19. The method of claim 15 , further including determining the first likelihood of detection based on at least one of a size of the reference shape or input from an administrator.
20. The method of claim 15 , further including determining the first likelihood of detection based on a Bayesian network.