Object detection and image cropping using a multi-detector approach

1. A computer-implemented method of detecting objects within digital video data, the method comprising:

defining an analysis profile comprising an initial number of analysis cycles dedicated to each of a plurality of detectors, each detector being independently configured to detect objects according to a unique set of analysis parameters and/or a unique detector algorithm;

receiving a plurality of frames of digital video data, the digital video data depicting at least one object;

analyzing the plurality of frames of digital video data using some or all of the plurality of detectors and in accordance with the analysis profile, wherein analyzing the plurality of frames produces an analysis result for each one of the plurality of detectors used in the analysis; and

updating the analysis profile by adjusting the number of analysis cycles dedicated to at least one of the plurality of detectors based on the analysis results.

2. The method as recited in claim 1 , wherein the digital video data depict a plurality of objects each having unique characteristics, and wherein the method comprises detecting at least two of the plurality of objects within one or more of the frames of digital video data.

3. The method as recited in claim 2 , wherein the plurality of frames of digital video data are consecutive frames.

4. The method as recited in claim 2 , wherein the detecting is performed using at least two of the detectors; and

wherein each of the at least two of the detectors exhibits a different analysis result with respect to the at least two detected objects.

5. The method as recited in claim 4 , wherein each of the different analysis results comprises:

a confidence score;

at least a portion of one or more of the frames of digital video data;

an image of at least a portion of one or more of the at least two detected objects; and/or

coordinates corresponding to a projected location, within one or more frames of the digital video data and/or the image, of one of the at least two detected objects.

6. The method as recited in claim 4 , wherein the adjusting comprises promoting at least one of the at least two detectors.

7. The method as recited in claim 6 , wherein the promoting increases the number of analysis cycles dedicated to one of the plurality of detectors exhibiting an optimal analysis result.

8. The method as recited in claim 4 , wherein the updating comprises demoting at least one of the at least two detectors.

9. The method as recited in claim 8 , wherein the demoting decreases the number of analysis cycles dedicated to one of the plurality of detectors exhibiting a least optimal analysis result.

10. The method as recited in claim 4 , wherein the detecting comprises performing a plurality of iterations of the analyzing and the updating.

11. The method as recited in claim 1 , wherein the plurality of detectors comprise at least one line segment detector.

12. The method as recited in claim 11 , comprising:

binarizing one or more frames of the digital video data using multiple binarization thresholds, wherein at least one of the binarization thresholds is adaptively determined based on analyzing the one or more frames of the digital video data;

generating an edge image based at least in part on a plurality of the binarized frames of the digital video data; and

detecting a plurality of line segments from within the edge image.

13. The method as recited in claim 12 , wherein detecting the plurality of line segments is based on gradient orientation and gradient magnitude information derived from the edge image.

14. The method as recited in claim 12 , comprising:

clustering some or all of the plurality of line segments; and

estimating a polygon corresponding to the at least one object based at least in part on the clustered plurality of line segments.

15. The method as recited in claim 1 , comprising selecting and/or parameterizing at least one of the plurality of detectors based at least in part on a source of the digital video data.

16. The method as recited in claim 1 , comprising projecting a location of the at least one object within the digital video data.

17. The method as recited in claim 1 , comprising combining analysis results from two or more of the plurality of detectors.

18. The method as recited in claim 17 , wherein each analysis result corresponds to a different portion of the at least one object.

19. A computer program product for detecting objects within digital video data, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions configured to cause a computer to perform a method comprising:

defining, using the computer, an analysis profile comprising an initial number of analysis cycles dedicated to each of a plurality of detectors, each detector being independently configured to detect objects according to a unique set of analysis parameters and/or a unique detector algorithm;

receiving, by the computer, a plurality of frames of digital video data, the digital video data depicting at least one object;

analyzing, using the computer, the plurality of frames of digital video data using some or all of the plurality of detectors and in accordance with the analysis profile, wherein analyzing the plurality of frames produces an analysis result for each used one of the plurality of detectors; and

updating, using the computer, the analysis profile by adjusting the number of analysis cycles dedicated to at least one of the plurality of detectors based on the analysis results.

20. A system for detecting objects within digital video data, the system comprising:

at least one processor; and

logic in and/or executable by the processor to cause the processor, upon execution thereof, to perform a method comprising:

defining, using the processor, an analysis profile comprising an initial number of analysis cycles dedicated to each of a plurality of detectors, each detector being independently configured to detect objects according to a unique set of analysis parameters and/or a unique detector algorithm;

receiving a plurality of frames of digital video data, the digital video data depicting at least one object;

analyzing, using the processor, the plurality of frames of digital video data using some or all of the plurality of detectors and in accordance with the analysis profile, wherein analyzing the plurality of frames produces an analysis result for each used one of the plurality of detectors; and

updating, using the processor, the analysis profile by adjusting the number of analysis cycles dedicated to at least one of the plurality of detectors based on the analysis results.