Uber Technologies, Inc.

Susmit Biswas

Kapil Gupta

Jeremy Hintz

Lionel Gueguen

Benjamin James Kadlec

Sai Bhuvan Teja Balla

Juthika Kiran Dabholkar

Houtan Shirani-Mehr

Chun-Chen Kuo

Roberto Fonti

Moustafa Elshikh

Systems and methods are provided for generating sets of candidates comprising images and places within a threshold geographic proximity based on geographic information associated with each of the plurality of images and geographic information associated with each place. For each set of candidates, the systems and methods generate a similarity score based on a similarity between text extracted from each image and a place name, and the geographic information associated with each image and each place. For each place with an associated image as a potential match, the systems and methods generate a name similarity score based on matching the extracted text of the image to the place name, and store an image as place data associated with a place based on determining that the name similarity score for the extracted text associated with the image is higher than a second predetermined threshold.

Imagery evidence matching system

Systems and methods are provided for receiving geographic coordinates for a location, determining a road segment associated with the location based on the geographic coordinates for the location, and determining a plurality of places associated with the road segment associated with the location. The systems and methods further provide for extracting visual data for each of the plurality of places, generating a plurality of feature values based on the visual data for each of the plurality of places, and analyzing the plurality of feature values to generate a visibility score for each of the plurality of places.

Place visibility scoring system

Systems and methods are provided for generating a places dataset comprising place data from multiple data sources each having a plurality of place data. Based on the places dataset, the systems and methods generate local clusters comprising places located within a predetermined distance, generate probable pairs of places in each local cluster, and distribute the probable pairs to one or more processors to determine similar edges between one or more places in each local cluster based on a plurality of combined similarity measures indicating whether or not a place is similar to another place. The systems and methods further generate similar pairs based on the combined similarity measures and convert the similar pairs into connected components comprising groups of similar places based on a best similar edge from all the candidates for a group of similar places.

Multiple source place matching system

Growing community of inventors

San Mateo, CA, United States of America

Susmit Biswas