Use natural language processing and multiple object detection models to automatically select objects in images

摘要

The present disclosure relates to an object selection system that automatically detects and selects objects in a digital image based on natural language inputs. For example, the object selection system may employ natural language processing tools to detect objects and their corresponding relationships within natural language object selection queries. The object selection system can, for example, determine alternative object terms for unrecognized objects in an object selection request based on natural language. In another example, the object selection system may determine multiple types of relationships between objects in a natural language object selection query and employ different object relationship models to select the requested query object.