Methods, apparatus and articles of manufacture to detect shapes

1. A method, comprising:

determining, by executing an instruction with a processor, a measured distance from an image capturing device to an object outline detected in image data representative of an environment, the determining of the measured distance including aggregating a plurality of depth values associated with the object outline;

when the measured distance corresponds to a first distance, selecting, by executing an instruction with the processor, a first group of reference shapes from a database for comparison to the object outline, the first group representing a subset of a collection of reference shapes in the database, the first group being selected to meet a first likelihood threshold of being detected at the first distance, the first likelihood threshold being different from a second likelihood threshold used to select a second group of reference shapes when the measured distance corresponds to a second distance different from the first distance; and

limiting comparison of the object outline to the reference shapes of the first group to reduce computational resource usage relative to comparing the object outline to each one of the collection of reference shapes in the database.

2. The method as defined in claim 1 , further including identifying an object corresponding to the detected object outline based on a similarity between the detected object outline and a first one of the reference shapes of the first group.

3. The method as defined in claim 2 , further including:

incrementing a people tally when the identified object corresponds to a person; and

associating the people tally with media detected in the environment at a same time as when the detection of the object outline occurred.

4. The method as defined in claim 1 , wherein the reference shapes of the first group are determined unlikely to be detectable at the second distance.

5. The method as defined in claim 1 , wherein the plurality of depth values are associated with points that fall within the detected object outline, the points selected according to a predetermined pattern.

6. A tangible machine readable storage medium comprising instructions that, when executed, cause a machine to at least:

determine a measured distance from an image capturing device to an object outline detected in image data representative of an environment, the measured distance determined by aggregating a plurality of depth values associated with the object outline;

when the measured distance corresponds to a first distance, selecting, for comparison to the object outline, a first subset of a collection of reference shapes from a database likely to be detected at the first measured distance, the first subset being selected to meet a first likelihood threshold of being detected at the first distance, the first likelihood threshold being different from a second likelihood threshold used to select a second subset of the collection of reference shapes when the measured distance corresponds to a second distance different from the first distance; and

limit comparison of the object outline to the reference shapes of the first subset to reduce computational resource usage relative to comparing the object outline to each one of the collection of reference shapes of the database.

7. The storage medium as defined in claim 6 , wherein the instructions, when executed, further cause the machine to identify an object corresponding to the object outline based on a similarity between the object outline and a first one of the reference shapes of the first subset.

8. The storage medium as defined in claim 7 , wherein the instructions, when executed, further cause the machine to:

increment a people tally when the object corresponds to a person; and

associate the people tally with media detected in the environment at a time corresponding to the detection of the object outline.

9. The storage medium as defined in claim 6 , wherein the first subset includes a reference shape unlikely to be detectable at the second distance.

10. The storage medium as defined in claim 6 , wherein the plurality of depth values are associated with points that fall within the detected object outline, the points selected according to a predetermined pattern.

11. An apparatus, comprising:

a calculator to calculate a first distance from an image capturing device to an object outline detected in image data representative of an environment by aggregating a plurality of depth values associated with the object outline;

a selector to select, based on the first distance, a first subset of reference shapes from a plurality of reference shapes in a database for comparison to the object outline, the first subset being selected to meet a first likelihood threshold of being detected at the first distance, the first likelihood threshold being different from a second likelihood threshold used to select a second subset of the collection of reference shapes when the measured distance corresponds to a second distance different from the first distance, at least one of the calculator or the selector being implemented via a logic circuit; and

a comparator to limit comparison of the object outline to the reference shapes of the first subset to reduce computational resource usage relative to comparing the object outline to each one of reference shapes in the database.

12. The apparatus as defined in claim 11 , wherein the comparator is to identify an object corresponding to the object outline based on a similarity between the object outline and a first one of the reference shapes of the first subset.

13. The apparatus as defined in claim 11 , wherein the plurality of depth values are associated with points that fall within the detected object outline, the points selected according to a predetermined pattern.

14. A method, comprising:

identifying, via a logic circuit, an object outline in image data generated by an image capturing device;

determining, via the logic circuit, a first distance between the object outline and the image capturing device by aggregating a plurality of depth values associated with the object outline;

identifying a first subset of a collection of reference shapes in a database, the first subset selected to meet a first likelihood threshold of being detected at the first distance, the first likelihood threshold being different from a second likelihood threshold used to select a second subset of the collection of reference shapes when the measured distance corresponds to a second distance different from the first distance; and

limiting comparison of the object shape to at least some of the first subset of the collection of reference shapes to reduce computational resource usage relative to comparing the object outline to each one of the collection of reference shapes in the database.

15. The method as defined in claim 14 , wherein the plurality of depth values are associated with points that fall within the detected object outline, the points selected according to a predetermined pattern.