Verifying a target object with reverse-parallax analysis

1. A method comprising:

a vehicle system detecting an object, the vehicle system being communicatively coupled with a vehicle;

the vehicle system simultaneously (i) capturing, via a first camera located at a first location, a first image of the detected object, the detected object having a position in the first image, and (ii) capturing, via a second camera located at a second location, a second image of the detected object;

the vehicle system retrieving location data specifying (a) a location of a target object, (b) the first location, (c) the second location, (d) a direction of the first camera, and (e) a direction of the second camera; and

based on the location data and the position of the detected object in the first image, the vehicle system predicting where in the second image the detected object would appear if the detected object is the target object.

2. The method of claim 1 , wherein the first location is a location of the first camera mounted on the vehicle, and wherein the second location is a location of the second camera mounted on the vehicle.

3. The method of claim 2 , wherein the first location and the second location are distinct locations.

4. The method of claim 1 , further comprising:

the vehicle system evaluating whether the detected object appears in a predetermined area of the second image, the predetermined area being based on the prediction of where in the second image the detected object would appear if the detected object is the target object.

5. The method of claim 4 , further comprising:

based on the evaluating, the vehicle system determining that the detected object does not appear in the predetermined area of the second image; and

in response to the determining, the vehicle system identifying the detected object as an object other than the target object.

6. The method of claim 5 , wherein the vehicle system predicting where in the second image the detected object would appear if the detected object is the target object comprises the vehicle system designating as a prediction zone a certain number of pixels of the second image, and

wherein evaluating whether the detected object appears in a predetermined area of the second image comprises the vehicle system determining whether a threshold number of pixels of the detected object appears within the prediction zone.

7. The method of claim 6 , further comprising:

based on the location data and a position of the detected object in the second image; the vehicle system predicting where in the first image the detected object would appear if the detected object is the target object.

8. The method of claim 7 , further comprising:

the vehicle system evaluating whether the detected object appears in a predetermined area of the first image, the predetermined area being based on the prediction of where in the first image the detected object would appear if the detected object is the target object.

9. The method of claim 1 , further comprising:

after predicting where in the second image the detected object would appear if the detected object is the target object, the vehicle system capturing, via a third camera located at a third location, a third image of the detected object;

the vehicle system retrieving additional location data specifying (a) the third location and (b) a direction of the third camera; and

based on the additional location data and further based on the predicting of where in the second image the detected object would appear if the detected object is the target object, the vehicle system predicting where in the third image the detected object would appear if the detected object is the target object.

10. The method of claim 9 , wherein the third location is distinct from one or both of the first location and the second location.

11. The method of claim 9 , wherein the third camera is the same camera as either the first camera or the second camera, and

wherein the third location is the same location as either the first location or the second location.

12. The method of claim 1 , further comprising:

after predicting where in the second image the detected object would appear if the detected object is the target object, based on the location data and further based on the predicting of where in the second image the detected object would appear if the detected object is the target object, the vehicle system (i) selecting a third location distinct from the first location and the second location and (ii) predicting where in a third image captured from the third location the detected object would appear if the detected object is the target object, wherein the predicting where in the third image the detected object would appear if the detected object is the target object occurs without having captured the third image from the third location; and

after selecting the third location and predicting where in the third image the detected object would appear if the detected object is the target object, the vehicle system causing the vehicle to move to the third location.

13. The method of claim 12 , wherein the vehicle system selecting a third location distinct from the first location and the second location comprises:

the vehicle system selecting a third location along a predetermined travel pattern of the vehicle.

14. The method of claim 12 , wherein the vehicle system selecting a third location distinct from the first location and the second location comprises:

the vehicle system selecting a third location that is not along a predetermined travel pattern of the vehicle.

15. The method of claim 1 , wherein the first location and the second location are along a predetermined travel pattern of the vehicle.

16. A vehicle comprising:

a vision system;

at least one processor; and

data storage comprising instructions executable by the at least one processor to cause the system to perform operations, the operations comprising:

causing the vision system to detect an object;

causing the vision system to simultaneously (i) capture, via a first camera located at a first location, a first image of the detected object, the detected object having a position in the first image, and (ii) capture, via a second camera located at a second location, a second image of the detected object;

retrieving location data specifying (a) a location of a target object, (b) the first location, (c) the second location, (d) a direction of the first camera, and (e) a direction of the second camera; and

based on the location data and the position of the detected object in the first image, predicting where in the second image the detected object would appear if the detected object is the target object.

17. The vehicle of claim 16 , the operations further comprising:

after predicting where in the second image the detected object would appear if the detected object is the target object, capturing, via a third camera located at a third location, a third image of the detected object;

retrieving additional location data specifying (a) the third location and (b) a direction of the third camera; and

based on the additional location data and further based on the predicting of where in the second image the detected object would appear if the detected object is the target object, predicting where in the third image the detected object would appear if the detected object is the target object.

18. A non-transitory computer readable medium (CRM) having stored thereon instructions executable by a computing device to cause the computing device to perform operations, the operations comprising:

detecting an object in an environment of a vehicle;

simultaneously (i) capturing, via a first camera located at a first location, a first image of the detected object, the detected object having a position in the first image, and (ii) capturing, via a second camera located at a second location, a second image of the detected object;

retrieving location data specifying (a) a location of a target object, (b) the first location, (c) the second location, (d) a direction of the first camera, and (e) a direction of the second camera; and

based on the location data and the position of the detected object in the first image, predicting where in the second image the detected object would appear if the detected object is the target object.

19. The CRM of claim 18 , wherein predicting where in the second image the detected object would appear if the detected object is the target object comprises designating as a prediction zone a certain number of pixels of the second image, and

wherein the operations further comprise:

evaluating whether the detected object appears in a predetermined area of the second image by determining whether a threshold number of pixels of the detected object appears within the prediction zone.

20. The CRM of claim 18 , the operations further comprising:

after predicting where in the second image the detected object would appear if the detected object is the target object, based on the location data and further based on the predicting of where in the second image the detected object would appear if the detected object is the target object, (i) selecting a third location distinct from the first location and the second location and (ii) predicting where in a third image captured from the third location the detected object would appear if the detected object is the target object, wherein the predicting where in the third image the detected object would appear if the detected object is the target object occurs without having captured the third image from the third location; and

after selecting the third location and predicting where in the third image the detected object would appear if the detected object is the target object, causing the vehicle to move to the third location.