Unified framework for analysis and recognition of identity documents

A system comprising:

• at least one hardware processor;
• and
• one or more software modules that are configured to, when executed by the at least one hardware processor,
• in each of a plurality of iterations of a document-recognition session,
• receive an image from a sequence of multiple input images captured from the same physical document,
• locate the document in the image and attempt to identify one or more of a plurality of templates that match the document,
• when one or more templates that match the document are identified, for each of the one or more templates,
• for each of a plurality of zones in the template, determine whether any of the plurality of zones in the template satisfy a zone-level stopping condition in which all objects in the zone have satisfied an object-level stopping condition, extract a sub-image of the zone from the image that has not satisfied the zone-level stopping condition, and skip a zone that has satisfied the zone-level stopping condition in one or more prior iterations,
• for each extracted sub-image, extract one or more objects from the sub-image of the zone that has not satisfied the object-level stopping condition, and skip an object that has satisfied the object-level stopping condition in one or more prior iterations, and,
• for each extracted object, perform object recognition on the object, and perform document recognition based on the one or more templates and results of the object recognition performed for each extracted object;
• accumulate, in non-persistent storage of the document-recognition session, a result of the document recognition performed in the iteration with a result of the document recognition performed in one or more prior iterations, and
• output a final result based on the accumulated result of the document recognition in the plurality of iterations.

A non-transitory computer-readable medium having instructions stored therein, wherein the instructions, when executed by a processor, cause the processor to:

• in each of a plurality of iterations of a document-recognition session,
• receive an image from a sequence of multiple input images captured from the same physical document, locate the document in the image and attempt to identify one or more of a plurality of templates that match the document,
• when one or more templates that match the document are identified, for each of the one or more templates,
• for each of a plurality of zones in the template, determine whether any of the plurality of zones in the template satisfy a zone-level stopping condition in which all objects in the zone have satisfied an object-level stopping condition, extract a sub-image of the zone from the image that has not satisfied the zone-level stopping condition, and skip a zone that has satisfied the zone-level stopping condition in one or more prior iterations,
• for each extracted sub-image, extract one or more objects from the sub-image of the zone that has not satisfied the object-level stopping condition, and skip an object that has satisfied the object-level stopping condition in one or more prior iterations, and,
• for each extracted object, perform object recognition on the object, and
• perform document recognition based on the one or more templates and results of the object recognition performed for each extracted object;
• accumulate, in non-persistent storage of the document-recognition session, a result of the document recognition performed in the iteration with a result of the document recognition performed in one or more prior iterations, and
• output a final result based on the accumulated result of the document recognition in the plurality of iterations.