Image processing for person recognition

A computer-implemented method at an image processing system, the method comprising:

• storing, at a memory, a plurality of images depicting a person unidentified by the image processing system;
• storing a trained probabilistic model that describes a relationship between image features using a context model;
• extracting, using a processor, features from an image of the plurality of images, one of the features comprising input sensed environmental data comprises:
• a time of day the image was captured;
• and
• a region of interest of the unidentified person;
• and
• computing a prediction of a name of the unidentified person depicted in the image using the extracted features and the context model.

A computer-implemented method at an image processing system, the computer-implemented method comprising:

• accessing an image depicting a person unidentified by an image processing system;
• extracting features from the image, the features comprising:
• input sensed environmental data comprising global positioning sensor data, and a time of day the image was captured;
• and a region of interest of the unidentified person;
• and
• accessing, from a trained probabilistic model, a description of a relationship between image features using a context model;
• and
• predicting a name of the unidentified person depicted in the image using the extracted features and the probabilistic model.