Visual monitoring of queues using auxillary devices

1. A method for monitoring a queue comprising:

receiving by one or more sensors a non-visual identifier within a queue;

capturing by one or more image capture devices images of an object possessing the non-visual identifier within the queue;

visually tracking another object in the queue based on transformations of at least one predefined feature extracted from the images of the object possessing the non-visual identifier; and

generating a feature transformation map to model changes of the at least one predefined feature at locations throughout the queue so that visual changes of the object possessing the nonvisual identifier are predictable and that the another object without possessing the non-visual identifier can be identified within the queue.

2. The method of claim 1 , wherein the transformations comprise models of changes in the at least one predefined feature between segments of the queue.

3. The method of claim 1 , wherein the locations throughout the queue include an entrance of the queue and an exit of the queue.

4. The method of claim 1 , wherein the visually tracking the another object is based on images of the another object at segments of the queue.

5. A system implemented in hardware for visually monitoring a queue, comprising:

a computing device communicatively linked to one or more non-visual sensors and one or more image capture devices, wherein the one or more computing devices:

detect an object at an entrance of the queue; and

based on a determination that the one or more non-visual sensors did not receive a non-visual identifier from the object, visually track the object in the queue using a feature transformation map, wherein

the feature transformation map models changes of a predefined feature of the object at locations throughout the queue so that visual changes in the object are predictable.

6. The system of claim 5 , wherein visually tracking the object in the queue comprises iteratively:

capturing a first image of the object at a location in the queue;

extracting a first plurality of features from the first image captured at the location; and

identifying the object using the first plurality of features, a second plurality of features extracted from another image captured at an adjacent location in the queue, and the feature transformation map.

7. The system of claim 5 , wherein the feature transformation map associates a plurality of adjacent locations in the queue with respective changes in a plurality of features extracted from images of a plurality of objects possessing non-visual identifiers.

8. The system of claim 7 , wherein the feature transformation map models visual changes in the plurality of predefined features for the adjacent locations of the queue.

9. A computer program product for monitoring a queue, the computer program product comprising:

one or more computer-readable, tangible storage devices, wherein the one or more computer-readable, tangible storage devices are not a transitory signal per se;

program instructions, stored on at least one of the one or more storage devices, that detect an object entering the queue;

program instructions, stored on at least one of the one or more storage devices, that determine whether a non-visual sensor received a non-visual identifier from the object;

program instructions, stored on at least one of the one or more storage devices, that visually track the object in the queue using a feature transformation map; and

program instructions, stored on at least one of the one or more storage devices, that predict visual changes in the object using the feature transformation map by modeling changes of a predefined feature associated with the object at locations throughout the queue so that another object can be identified and localized within the queue.

10. The computer program product of claim 9 , wherein the feature transformation map is a data structure associating each location of the grid with one or more adjacent locations of the grid and corresponding changes in visual features between the associated locations and adjacent locations.

11. The computer program product of claim 9 , further comprising program instructions stored on at least one of the one or more storage devices that determine a waiting time of the queue.