Systems and method for visual search with attribute manipulation

A computer-implemented method, comprising:

• obtaining an original feature vector representation of an item, the original feature vector having a first plurality of values corresponding to a plurality of attributes;
• selecting an attribute of the plurality of attributes to modify based on a preprogrammed protocol;
• modifying a value associated with the attribute based at least in part on the preprogrammed protocol;
• generating a target feature vector having a second plurality of values corresponding to at least some of the plurality of attributes,
• wherein the target feature vector has the same values as the original feature vector for at least a first attribute of the plurality of attributes;
• and
• wherein the target feature vector has different values from the original feature vector or no values for at least a second attribute of the plurality of attributes;
• querying a database of items using the target feature vector, wherein the database includes a plurality of images associated with a plurality of items, respectively;
• comparing the target feature vector to feature vector representations of the plurality of images;
• and
• determining one or more candidate items from the database based at least in part on similarity between the feature vector representation of images associated with the one or more candidate items and the target feature vector.

A system, comprising:

• at least one computing device processor;
• and
• a memory device including instructions that, when executed by the at least one computing device processor, cause the system to:
• obtain an original feature vector representation of an item, the original feature vector having a first plurality of values corresponding to a plurality of attributes;
• select an attribute of the plurality of attributes to modify based on a preprogrammed protocol;
• modify a value associated with the attribute based at least in part on the preprogrammed protocol;
• generate a target feature vector having a second plurality of values corresponding to at least some of the plurality of attributes,
• wherein the target feature vector has the same values as the original feature vector for at least a first attribute of the plurality of attributes;
• and
• wherein the target feature vector has different values from the original feature vector or no values for at least a second attribute of the plurality of attributes;
• query a database of items using the target feature vector, wherein the database includes a plurality of images associated with a plurality of items, respectively;
• compare the target feature vector to feature vector representations of the plurality of images;
• and
• determine one or more candidate items from the database based at least in part on similarity between the feature vector representation of images associated with the one or more candidate items and the target feature vector.