Object recognition and presentation for the visually impaired

A method to perform object recognition, the method comprising:

• capturing, by a camera, a plurality of images of a field of view of the camera, the camera in communication with processing circuitry;
• repeatedly comparing, by the processing circuitry, characteristic features within images of the plurality of images to an object identification dataset to identify object matches for a plurality of objects within the plurality of images;
• determining, by the processing circuitry, identified objects based on object matches in the plurality of images;
• filtering, by the processing circuitry, between iterations of comparing the characteristic features to the object identification dataset, to consecutive images over time to estimate positions of the identified objects based on a prior location determination of the identified objects to represent movement with an estimated track state prior to performing object recognition;
• and
• transmitting, by the processing circuitry, a position of an identified object based on the estimated track state.