US 2026/0288501 A1Published application

Page-registered contextual reading assistance for physical printed text

OwnerPrithvi Karthikeyan Mittalinventor

Life of this patent

Estimating the term…

Loading patent…

Abstract

A reading assistance device, system, and method operate on live imagery of a physical printed page to identify selected text and provide contextual explanations. Optical character recognition performed on the live imagery generates recognized text tokens with positional metadata within a page coordinate system. A reader interaction signal associated with a selected region of the physical printed page is mapped into the page coordinate system to resolve a target token or phrase. Surrounding recognized text is processed to determine meaning within the printed material. Contextual output is generated based on the determined meaning and, in certain embodiments, reader context corresponding to age level, reading stage, or proficiency. The contextual output is rendered in spatial association with the selected text while the underlying printed text remains visible. The system operates without requiring retrieval of a pre-existing electronic copy.

Claims20

Dependent claims

A reading assistance device for interacting with physical printed text, the device comprising:

an imaging sensor configured to capture live imagery of a physical printed page containing printed text;
a processor;
memory operatively coupled to the processor;
and
a display positioned relative to the physical printed page and configured to present visual output in page-registered relation to the physical printed page,
wherein the memory stores instructions that, when executed by the processor, cause the device to:
capture live imagery of the physical printed page using the imaging sensor;
establish or access a page coordinate system corresponding to the physical printed page;
perform optical character recognition on the live imagery to generate a plurality of recognized text tokens, each recognized text token being associated with positional metadata within the page coordinate system;
receive a selection signal associated with a selected region of the physical printed page;
map the selected region into the page coordinate system;
determine, based on a positional relationship between the selected region and the positional metadata of the recognized text tokens, a target token or phrase on the physical printed page from the recognized text tokens generated from the live imagery, without requiring retrieval of a pre-existing electronic copy of the physical printed page;
extract contextual text associated with the target token or phrase from the recognized text tokens;
generate contextual output comprising textual explanatory content and optionally visual explanatory content and/or auditory explanatory content;
and
render, by the display, the contextual output at a display position determined from the positional metadata of the target token or phrase such that the contextual output is page-registered relative to the physical printed page while the underlying printed text remains visible.

13 more elements

A method of providing contextual reading assistance for physical printed text, the method comprising:

capturing live imagery of a physical printed page containing printed text;
performing optical character recognition directly on the live imagery to generate recognized text tokens each associated with positional metadata including spatial coordinates, without requiring retrieval of a pre-existing electronic copy of the physical printed page;
generating a page coordinate system corresponding to the physical printed page;
determining a coordinate transformation between image coordinates and the page coordinate system;
receiving a reader interaction signal associated with a selected region of the physical printed page;
mapping the selected region into the page coordinate system;
resolving, based on a positional relationship between the selected region and the spatial coordinates of the recognized text tokens, a target token or phrase;
extracting contextual text associated with the target token or phrase;
generating contextual output;
and
rendering the contextual output in page-registered relation to the target token or phrase on the physical printed page while the underlying printed text remains visible.

8 more elements

Dependencies are inferred from the claim text.