| CPC G06F 16/532 (2019.01) | 20 Claims |
|
1. A computer-implemented method comprising:
extracting, by a computing system comprising one or more processor devices, one or more visual descriptors for a particular Point of Interest (POI) from a corpus of User-Generated Content (UGC), wherein the corpus of UGC comprises a plurality of textual content items descriptive of the particular POI, and wherein at least one of the one or more visual descriptors are extracted from the plurality of textual content items;
based on the one or more visual descriptors, using, by the computing system, a machine-learned generative vision model to generate a query image that depicts the particular POI;
determining, by the computing system, a visual similarity between the query image and each of a plurality of candidate images of a corpus of visual UGC that depicts the particular POI; and
selecting, by the computing system, a representation image from the plurality of candidate images to represent the particular POI based on the visual similarity between the representation image and the query image.
|