User-specific landmarks for navigation systems

1. A computer-implemented method comprising:

performing a learning phase updating a data store based on previous trips of drivers to associate drivers with landmarks from the previous trips of the drivers, the learning phase comprising:

receiving, from a first rider, trip location data comprising a pickup location and a destination location;

storing, in the data store, trip data describing a trip from the pickup location to the destination location in association with the first rider and a driver;

based on the pickup location or the destination location, identifying landmark information of a landmark;

storing the identified landmark information in the data store in association with the driver;

performing a usage phase using the association between the identified landmark information and the driver to provide to the driver driving directions that reference the landmark, the usage phase comprising:

receiving, from a second rider, second trip location data comprising a second pickup location and a second destination location;

receiving, from the driver, trip position data indicating a current location of the driver;

based on the current location, identifying a route from the current location toward the second destination location;

determining that the landmark is on the route;

responsive to determining that the landmark is on the route;

determining whether the landmark is visible from a street location; and

responsive to determining that the landmark is visible from the street location:

causing display of an indication of the landmark in a map user interface of the driver; and

providing a turn direction to the driver that references the landmark.

2. The computer-implemented method of claim 1 , further comprising, for a second driver different from the driver, and for the route, refraining from causing display of the indication of the landmark in a user interface of the second driver responsive to the second driver not being associated with the landmark.

3. The computer-implemented method of claim 1 , wherein identifying the landmark information comprises searching within map data to identify landmarks proximate to the pickup location or to the destination location.

4. The computer-implemented method of claim 1 , further comprising computing a significance score for the landmark based on a number of times that the driver has visited the landmark.

5. The computer-implemented method of claim 1 , further comprising:

computing a significance score for the landmark; and

refraining from causing display of an indication of a second landmark having a lower significance score than the significance score for the landmark, responsive to the second landmark being located proximate to the landmark.

6. The computer-implemented method of claim 1 , further comprising computing a significance score for the landmark using visit context data describing a purchase made by the driver at the landmark.

7. The computer-implemented method of claim 1 , further comprising:

computing a significance score for the landmark: and

responsive to the landmark having a significance score of at least a threshold value, displaying the landmark with a visual emphasis in the user interface of the driver.

8. The computer-implemented method of claim 1 , wherein determining whether the landmark is visible from the street location comprises:

obtaining street image data for a street location on the route; and

deriving text corresponding to the landmark within the street image data using optical character recognition.

9. The computer-implemented method of claim 1 , wherein determining whether the landmark is visible from the street location comprises:

obtaining street image data for a street location on the route; and

performing image analysis of the street image data to recognize the landmark within the street image data.

10. A non-transitory computer-readable storage medium storing executable instructions that when executed by a processor perform actions comprising:

performing a learning phase updating a data store based on previous trips of drivers to associate drivers with landmarks from the previous trips of the drivers, the learning phase comprising:

receiving, from a first rider, trip location data comprising a pickup location and a destination location;

storing, in the data store, trip data describing a trip from the pickup location to the destination location in association with the first rider and a driver;

based on the pickup location or the destination location, identifying landmark information of a landmark;

storing the identified landmark information in the data store in association with the driver;

performing a usage phase using the association between the identified landmark information and the driver to provide to the driver driving directions that reference the landmark, the usage phase comprising:

receiving, from a second rider, second trip location data comprising a second pickup location and a second destination location;

receiving, from the driver, trip position data indicating a current location of the driver;

based on the current location, identifying a route from the current location toward the second destination location;

determining that the landmark is on the route;

responsive to determining that the landmark is on the route:

determining whether the landmark is visible from a street location; and

responsive to determining that the landmark is visible from the street location:

causing display of an indication of the landmark in a map user interface of the driver; and

providing a turn direction to the driver that references the landmark.

11. The non-transitory computer-readable storage medium of claim 10 , the actions further comprising, for a second driver different from the driver, and for the route, refraining from causing display of the indication of the landmark in a user interface of the second driver responsive to the second driver not being associated with the landmark.

12. The non-transitory computer-readable storage medium of claim 10 , wherein identifying the landmark information comprises searching within map data to identify landmarks proximate to the pickup location or to the destination location.

13. The non-transitory computer-readable storage medium of claim 10 , the actions further comprising computing a significance score for the landmark based on a number of times that the driver has visited the landmark.

14. The non-transitory computer-readable storage medium of claim 10 , the actions further comprising:

computing a significance score for the landmark; and

refraining from causing display of an indication of a second landmark having a lower significance score than the significance score for the landmark, responsive to the second landmark being located proximate to the landmark.

15. The non-transitory computer-readable storage medium of claim 10 , the actions further comprising computing a significance score for the landmark using visit context data describing a purchase made by the driver at the landmark.

16. The non-transitory computer-readable storage medium of claim 10 , the actions further comprising:

computing a significance score for the landmark; and

responsive to the landmark having a significance score of at least a threshold value, displaying the landmark with a visual emphasis in the user interface of the driver.

17. A computer system comprising:

a computer processor; and

a computer-readable storage medium storing executable instructions that when executed by the computer processor perform actions comprising:

performing a learning phase updating a data store based on previous trips of drivers to associate drivers with landmarks from the previous trips of the drivers, the learning phase comprising:

receiving, from a first rider, trip location data comprising a pickup location and a destination location;

storing, in the data store, trip data describing a trip from the pickup location to the destination location in association with the first rider and a driver;

based on the pickup location or the destination location, identifying landmark information of a landmark;

storing the identified landmark information in the data store in association with the driver;

performing a usage phase using the association between the identified landmark information and the driver to provide to the driver driving directions that reference the landmark, the usage phase comprising:

receiving, from a second rider, second trip location data comprising a second pickup location and a second destination location;

receiving, from the driver, trip position data indicating a current location of the driver;

based on the current location, identifying a route from the current location toward the second destination location;

determining that the landmark is on the route;

responsive to determining that the landmark is on the route:

determining whether the landmark is visible from a street location; and

responsive to determining that the landmark is visible from the street location:

causing display of an indication of the landmark in a map user interface of the driver; and

providing a turn direction to the driver that references the landmark.

18. The computer system of claim 17 , the actions further comprising, for a second driver different from the driver, and for the route, refraining from causing display of the indication of the landmark in a user interface of the second driver responsive to the second driver not being associated with the landmark.

19. The computer system of claim 17 , wherein identifying the landmark information comprises searching within map data to identify landmarks proximate to the pickup location or to the destination location.

20. The computer system of claim 17 , the actions further comprising computing a significance score for the landmark based on a number of times that the driver has visited the landmark.

21. The computer system of claim 17 , the actions further comprising:

computing a significance score for the landmark; and

refraining from causing display of an indication of a second landmark having a lower significance score than the significance score for the landmark, responsive to the second landmark being located proximate to the landmark.

22. The computer system of claim 17 , the actions further comprising computing a significance score for the landmark using visit context data describing a purchase made by the driver at the landmark.