US 12,072,928 B2
Finding locally prominent semantic features for navigation and geocoding
Yan Mayster, Aurora, CO (US); Brian Brewington, Superior, CO (US); and Matthew Strosnick, Mountain View, CA (US)
Assigned to GOOGLE LLC, Mountain View, CA (US)
Appl. No. 16/977,777
Filed by Google LLC, Mountain View, CA (US)
PCT Filed Oct. 22, 2018, PCT No. PCT/US2018/056854
§ 371(c)(1), (2) Date Sep. 2, 2020,
PCT Pub. No. WO2020/086051, PCT Pub. Date Apr. 30, 2020.
Prior Publication US 2021/0240762 A1, Aug. 5, 2021
Int. Cl. G06F 16/9535 (2019.01); G06F 16/583 (2019.01); G06F 16/587 (2019.01); G06F 16/951 (2019.01); G06V 20/56 (2022.01)
CPC G06F 16/587 (2019.01) [G06F 16/583 (2019.01); G06V 20/56 (2022.01)] 18 Claims
OG exemplary drawing
 
1. A computer-implemented method of providing navigational instructions that reference landmarks, the method comprising:
accessing, by a computing system comprising one or more processors, a plurality of semantic tags associated with a plurality of images, wherein each of the plurality of semantic tags is associated with one or more features depicted by one of the plurality of images, and wherein each of the one or more features is associated with a geographic location;
identifying, by the computing system, based at least in part on the plurality of semantic tags, one or more landmarks comprising the one or more features that satisfy one or more entropic criteria that measure a localized prominence of each of the one or more features, wherein the one or more entropic criteria comprise a comparison of a rate at which the one or more features occur to a lower threshold rate, wherein the rate at which the one or more features occur is at a rate below the lower threshold rate and is determined based on a number of occurrences of the feature within a distance of a navigational travel path generated by a navigational computing system;
selecting, by the computing system, based at least in part on context data associated with a location on the navigational travel path generated by the navigational computing system comprising a plurality of locations and a level of familiarity with the one or more landmarks, at least one landmark for use in navigation at the location, wherein the level of familiarity is based at least in part on a user's previous association with the one or more features of the one or more landmarks; and
generating, by the computing system, at least one navigational instruction with associated text that references the at least one landmark.