Systems and methods for visual object matching

A non-transitory computer readable medium having stored therein instructions executable by a computing device to cause the computing device to perform functions of:

• receiving a plurality of query images;
• matching the plurality of query images to a plurality of objects using a visual object recognition module, wherein the visual object recognition module is configured to utilize information from a training corpus to match query images to one or more training images of the plurality of objects, wherein the training corpus comprises one or more training images associated with the plurality of objects;
• associating match scores with matches between the query images and one or more matched training images, wherein the visual object recognition module is configured to determine match scores indicating a level of similarity between the query images and one or more matched training images;
• determining matched and unmatched query images based on the match scores associated with the query images, wherein query images with match scores above a score threshold are identified as matched query images and query images with match scores below the score threshold are identified as unmatched query images;
• and
• based on the determination, expanding the training corpus utilized by the visual object recognition module to include the matched query images, wherein the matched query images are added as additional training images associated with the objects to which the matched training images are associated, and the matched query images are used as training images when re-training the visual object recognition module on the expanded training corpus.

A system comprising:

• a memory;
• a processor coupled to the memory;
• and
• instructions stored in the memory and executable by the processor to perform functions of:
• receiving a plurality of query images;
• matching the plurality of query images to a plurality of objects using a visual object recognition module, wherein the visual object recognition module is configured to utilize information from a training corpus to match query images to one or more training images of the plurality of objects, wherein the training corpus comprises one or more training images associated with the plurality of objects;
• associating match scores with matches between the query images and one or more matched training images, wherein the visual object recognition module is configured to determine match scores indicating a level of similarity between the query images and one or more matched training images;
• determining matched and unmatched query images based on the match scores associated with the query images, wherein query images with match scores above a score threshold are identified as matched query images and query images with match scores below the score threshold are identified as unmatched query images;
• and
• based on the determination, expanding the training corpus utilized by the visual object recognition module to include the matched query images, wherein the matched query images are added as additional training images associated with the objects to which the matched training images are associated, and the matched query images are used as training images when re-training the visual object recognition module on the expanded training corpus.