Utilizing natural language processing automatically select objects in images

A method comprising:

• analyzing a query string to identify an object term that indicates a query object to be selected in a digital image;
• determining that the object term does not correspond to a known object class;
• based on the object term not corresponding to a known object class, mapping the object term to an alternative object term corresponding to a first known object class;
• generating an object mask for the query object utilizing a known object detection neural network trained to identify objects of the first known object class;
• and
• providing the digital image with the object mask for the query object in response to receiving the query string.

A system comprising:

• one or more memory devices;
• and
• one or more server devices coupled to the one or more memory devices configured to cause the system to:
• receive a query string to select a query object in a digital image;
• analyze the query string to identify a plurality of object terms and at least one relationship term linking the plurality of object terms;
• identify a plurality of object classes corresponding to the plurality of object terms;
• generate object masks for objects of the plurality of object classes utilizing one or more object detection models;
• analyze the object masks generated for the objects of the plurality of object classes to identify an object mask that satisfies a relationship type defined by the at least one relationship term, by determining at least one of a spatial relationship, a proximity relationship, a depth relationship, a relative position relationship, an absolute position relationship, or an exclusion relationship between two or more of the object masks;
• and
• provide the digital image with the query object selected by providing the object mask in response to receiving the query string.