A method can include identifying at least one radical of a plurality of East Asian characters based at least in part on stroke information received via a touch-sensitive display; ranking individual East Asian characters of the plurality of East Asian characters according to their individual frequencies in a corpus; and rendering at least a portion of the ranked individual East Asian characters to the touch-sensitive display.

Determination of an underlying grid structure that facilitates layout of East Asian text is disclosed. The underlying grid structure includes both a size of character frames and a size of a text block frame. The East Asian text may be obtained from a scan of printed material that has the text formatted according to layout conventions established by the publisher. The text may be reformatted to appear on a display of an electronic device in a manner similar to the formatting in the original scanned document. Reformatting may include reflowing the text in order to fit a greater or lesser number of characters on a line. The reflowing may maintain character spacing from the original document and follow formatting rules against locating certain characters at the start or end of a line.

An information processing apparatus includes a processor configured to acquire each recognition result output by each of plural different recognition processes for the same image, and execute, in relation to the recognition result selected from among the recognition results output by each of the plural recognition processes, a postprocess corresponding to the recognition process from which the selected recognition result is output.

Techniques presented herein describe rotating text content rendered on a mobile device to a preferred orientation based on the mobile device orientation and language of the text content. A display tool executing on the mobile device detects an orientation of the mobile device changing from a first orientation to a second orientation. The mobile device currently displays the characters in the text content according to the first orientation. The display tool determines that the text content is written in a specified language. The display tool re-renders the display of the characters in the text content and any accompanying images according to the second orientation.

A method and apparatus of open set recognition, and a computer-readable storage medium are disclosed. The method comprises acquiring auxiliary data and training data of known categories for open set recognition, training a neural network alternately using the auxiliary data and the training data, until convergence; extracting a feature of data to be recognized for open set recognition, using the trained neural network; and recognizing a category of data to be recognized, based on the feature of the data to be recognized.

The present disclosure discloses a method for training an adversarial network model, a method for building a character library, an electronic device and a storage medium, which relate to a field of artificial intelligence, in particular to a field of computer vision and deep learning technologies, and are applicable in a scene of image processing and image recognition. The method for training includes: generating a new character by using the generation model based on a stroke character sample and a line character sample; discriminating a reality of the generated new character by using the discrimination model; calculating a basic loss based on the new character and a discrimination result; calculating a track consistency loss based on a track consistency between the line character sample and the new character; and adjusting a parameter of the generation model according to the basic loss and the track consistency loss.

Disclosed are systems, computer-readable mediums, and methods for determining that text contains Chinese, Japanese, or Korean characters. One method includes determining a language hypothesis for each text fragment in a plurality of text fragments identified from connected components in a document image. The method further includes selecting a first subset of text fragments from the plurality of text fragments based on ratings for the language hypothesis of each text fragment in the plurality of text fragments. The method further includes verifying, by a processor, the language hypothesis of one or more text fragments in the first subset of text fragments based on optical character recognition of the one or more text fragments. The method further includes determining, by the processor, that Chinese, Japanese, or Korean (CJK) characters are present in the document image based on the verification of the language hypothesis of each of the one or more text fragments.

Ideogram character analysis includes partitioning an original ideogram character into strokes, and mapping each stroke to a corresponding stroke identifier (id) to create an original stroke id sequence that includes stroke identifiers. A candidate ideogram character that has a candidate stroke id sequence within a threshold distance to the original stroke id sequence is selected. One or more embodiments may create new phrase by replacing the original ideogram character with the candidate ideogram character in a search phrase. One or more embodiments perform a search using the search phrase and the new phrase to obtain a result, and present the result. One or more embodiments may replace an original ideogram character in a character recognized document with the candidate ideogram character and store the character recognized document.

Provided are a method for training a font generation model, a method for establishing a font library, and a device. The method for training a font generation model includes the following steps: a source-domain sample character is input into the font generation model to obtain a first target-domain generated character; the first target-domain generated character and a preset target-domain sample character are input into a character classification model to obtain a first feature loss of the font generation model; the first target-domain generated character and the target-domain sample character are input into a font classification model to obtain a second feature loss of the font generation model; a target feature loss is determined according to the first feature loss and/or the second feature loss; and the model parameter of the font generation model is updated according to the target feature loss.

A feature extraction system, method and apparatus based on neural network optimization by gradient filtering is provided. The feature extraction method includes: acquiring, by an information acquisition device, input information; constructing, by a feature extraction device, different feature extraction networks, performing iterative training on the networks in combination with corresponding training task queues to obtain optimized feature extraction networks for different input information, and calling a corresponding optimized feature extraction network to perform feature extraction according to a class of the input information; performing, by an online updating device, online updating of the networks; and outputting, by a feature output device, a feature of the input information. The new feature extraction system, method and apparatus avoids the problem of catastrophic forgetting of the artificial neural network in continuous tasks, and achieves high accuracy and precision in continuous feature extraction.