Dynamic learning for object tracking
Techniques described herein relate to mobile computing device technologies, such as systems, methods, apparatuses, and computer-readable media for tracking an object from a plurality of objects. In one aspect, the plurality of objects may be similar. Techniques discussed herein propose dynamically learning information associated with each of the objects and discriminating between objects based on their differentiating features. In one implementation, this may be done by maintaining a database associated with each object and updating the dynamic database transferred while the objects are tracked. The tracker uses algorithmic means for differentiating objects by focusing on the differences amongst the objects. For example, in one implementation, the method may weigh the differences between different fingers higher than their associated similarities to facilitate differentiating the fingers.
1. A method for tracking an object, comprising:
detecting a plurality of objects from image data, wherein the plurality of objects comprises a first object, and wherein the first object is identified using a first label for a positive tracking object and remaining objects of the plurality of objects are identified using a second label for negative tracking objects to indicate that the first object is to be the focus of tracking and that the remaining objects are not to be the focus of tracking, and to indicate that information associated with the first object is maintained in a dataset different than information associated with the remaining objects of the plurality of objects;
tracking a difference associated with one or more features for the plurality of objects with respect to the first object from the image data, wherein tracking the difference associated with the one or more features for the plurality of objects comprises dynamically learning and storing information about the difference of the one or more features associated with one or more of the plurality of objects, and wherein the dynamically learned information is maintained in an IDFL (Interactive Decision Forest Learning) dataset for each detected object; and
identifying the first object from the plurality of objects using the tracking difference associated with the one or more features for the plurality of objects.
2. The method of claim 1 , wherein identifying the first object from the plurality of objects comprises weighing the difference associated with the one or more features amongst the plurality of objects higher than similarities associated with the one or more features amongst the plurality of objects.
3. The method of claim 1 , wherein tracking the difference associated with the one or more features comprises:
identifying the one or more features for each object from the plurality of objects; and
determining the difference associated with the one or more features between the first object and the remaining one or more objects.
4. The method of claim 3 , wherein the one or more features for each object are tracked using an instance of a tracker and wherein the difference associated with the one or more features between the first object and the remaining one or more objects is determined by sharing information between the instances of trackers associated with each object.
5. The method of claim 1 , wherein the plurality of objects are fingers.
6. The method of claim 1 , wherein differentiating the first object from at least one of the remaining one or more objects comprises using the difference in an at least one feature from the one or more features associated with the positive tracking object and the at least one of the remaining one or more objects labeled as negative tracking objects in differentiating the first object from the at least one of the remaining one or more objects.
7. The method of claim 1 , wherein the plurality of objects are substantially similar to each other.
8. The method of claim 1 , wherein detecting the plurality of objects further comprises determining the first object from the plurality of objects.
9. The method of claim 1 , further comprising identifying the first object using trajectory information associated with a movement of the first object.
10. The method of claim 9 , further comprising identifying the first object using the difference between the trajectory information associated with the movement of the first object and at least one of the remaining one or more objects.
11. The method of claim 1 , further comprising identifying the first object based on physical constraints on a movement of the first object.
12. A computing device, comprising:
a camera coupled to the computing device for acquiring an image;
a processor configured to:
detect a plurality of objects from image data from the image, wherein the plurality of objects comprises a first object, and wherein the first object is identified using a first label for a positive tracking object and remaining objects of the plurality of objects are identified using a second label for negative tracking objects to indicate that the first object is to be the focus of tracking and that the remaining objects are not to be the focus of tracking, and to indicate that information associated with the first object is maintained in a dataset different than information associated with the remaining objects of the plurality of objects; and
a tracking module executing on the processor configured to:
track a difference associated with one or more features for the plurality of objects with respect to the first object from the image data; and
identify the first object from the plurality of objects using the tracking difference associated with the one or more features for the plurality of objects.
13. The computing device of claim 12 , wherein tracking the difference associated with the one or more features, by the tracking module, for the plurality of objects comprises dynamically learning and storing information about the difference of the one or more features associated with one or more of the plurality of objects.
14. The computing device of claim 13 , wherein the dynamically learned information is maintained by the computing device in a dynamic decision forest learning dataset associated with each detected object.
15. The computing device of claim 12 , wherein identifying the first object from the plurality of objects, by the tracking module of the computing device, comprises weighing the difference associated with the one or more features amongst the plurality of objects higher than similarities associated with the one or more features amongst the plurality of objects.
16. The computing device of claim 12 , wherein tracking the difference associated with the one or more features comprises:
identifying, by the tracking module, the one or more features for each object from the plurality of objects; and
determining, by the tracking module, the difference associated with the one or more features between the first object and the remaining one or more objects.
17. The computing device of claim 16 , wherein the one or more features for each object are tracked, by the computing device, using an instance of a tracker and wherein the difference associated with the one or more features between the first object and the remaining one or more objects is determined by sharing information between the instances of trackers associated with each object.
18. The computing device of claim 12 , wherein the plurality of objects are fingers.
19. The computing device of claim 12 , wherein differentiating the first object from at least one of the remaining one or more objects comprises using the difference in an at least one feature from the one or more features associated with the positive tracking object and the at least one of the remaining one or more objects labeled as negative tracking objects in differentiating the first object from the at least one of the remaining one or more objects.
20. The computing device of claim 12 , wherein the plurality of objects are substantially similar to each other.
21. The computing device of claim 12 , wherein detecting the plurality of objects further comprises determining the first object from the plurality of objects.
22. The computing device of claim 12 , further comprising identifying the first object using trajectory information associated with a movement of the first object.
23. The computing device of claim 22 , further comprising identifying the first object using the difference between the trajectory information associated with the movement of the first object and at least one of the remaining one or more objects.
24. The computing device of claim 12 , further comprising identifying the first object based on physical constraints on a movement of the first object.
25. A non-transitory computer readable storage medium, wherein the non-transitory computer readable storage medium comprises instructions executable by a processor, the instructions comprising instructions to:
detect a plurality of objects from image data, wherein the plurality of objects comprises a first object, and wherein the first object is identified using a first label for a positive tracking object and remaining objects of the plurality of objects are identified using a second label for negative tracking objects to indicate that the first object is to be the focus of tracking and that the remaining objects are not to be the focus of tracking, and to indicate that information associated with the first object is maintained in a dataset different than information associated with the remaining objects of the plurality of objects;
track a difference associated with one or more features for the plurality of objects with respect to the first object from the image data; and
identify the first object from the plurality of objects using the tracking difference associated with the one or more features for the plurality of objects.
26. The non-transitory computer readable storage medium of claim 25 , wherein tracking the difference associated with the one or more features for the plurality of objects comprises dynamically learning and storing information about the difference of the one or more features associated with one or more of the plurality of objects.
27. The non-transitory computer readable storage medium of claim 25 , wherein identifying the first object from the plurality of objects comprises weighing the difference associated with the one or more features amongst the plurality of objects higher than similarities associated with the one or more features amongst the plurality of objects.
28. The non-transitory computer readable storage medium of claim 25 , wherein tracking the difference associated with the one or more features comprises instructions to:
identify the one or more features for each object from the plurality of objects; and
determine the difference associated with the one or more features between the first object and the remaining one or more objects.
29. The non-transitory computer readable storage medium of claim 28 , wherein the one or more features for each object are tracked using an instance of a tracker and wherein the difference associated with the one or more features between the first object and the remaining one or more objects is determined by sharing information between the instances of trackers associated with each object.
30. The non-transitory computer readable storage medium of claim 25 , wherein the plurality of objects are fingers.
31. The non-transitory computer readable storage medium of claim 25 , wherein the plurality of objects are substantially similar to each other.
32. An apparatus, comprising:
means for detecting a plurality of objects from image data, wherein the plurality of objects comprises a first object, and wherein the first object is identified using a first label for a positive tracking object and remaining objects of the plurality of objects are identified using a second label for negative tracking objects to indicate that the first object is to be the focus of tracking and that the remaining objects are not to be the focus of tracking, and to indicate that information associated with the first object is maintained in a dataset different than information associated with the remaining objects of the plurality of objects;
means for tracking a difference associated with one or more features for the plurality of objects with respect to the first object from the image data; and
means for identifying the first object from the plurality of objects using the tracking difference associated with the one or more features for the plurality of objects.
33. The apparatus of claim 32 , wherein tracking the difference associated with the one or more features for the plurality of objects comprises means for dynamically learning and storing information about the difference of the one or more features associated with the one or more of the plurality of objects.
34. The apparatus of claim 32 , wherein identifying the first object from the plurality of objects comprises means for weighing the difference associated with the one or more features amongst the plurality of objects higher than similarities associated with the one or more features amongst the plurality of objects.
35. The apparatus of claim 32 , wherein tracking the difference associated with the one or more features comprises instructions to:
means for identifying the one or more features for each object from the plurality of objects; and
means for determining the difference associated with the one or more features between the first object and the remaining one or more objects.
36. The apparatus of claim 35 , wherein the one or more features for each object are tracked using an instance of a tracker and wherein the difference associated with the one or more features between the first object and the remaining one or more objects is determined by sharing information between the instances of trackers associated with each object.
37. The apparatus of claim 32 , wherein the plurality of objects are fingers.
38. The apparatus of claim 32 , wherein the plurality of objects are substantially similar to each other.