摘要:
Tensor linear Laplacian discrimination for feature extraction is disclosed. One embodiment comprises generating a contextual distance based sample weight and class weight, calculating a within-class scatter using the at least one sample weight and a between-class scatter for multiple classes of data samples in a sample set using the class weight, performing a mode-k matrix unfolding on scatters and generating at least one orthogonal projection matrix.
摘要:
A method for modeling data affinities and data structures. In one implementation, a contextual distance may be calculated between a selected data point in a data sample and a data point in a contextual set of the selected data point. The contextual set may include the selected data point and one or more data points in the neighborhood of the selected data point. The contextual distance may be the difference between the selected data point's contribution to the integrity of the geometric structure of the contextual set and the data point's contribution to the integrity of the geometric structure of the contextual set. The process may be repeated for each data point in the contextual set of the selected data point. The process may be repeated for each selected data point in the data sample. A digraph may be created using a plurality of contextual distances generated by the process.
摘要:
A method for modeling data affinities and data structures. In one implementation, a contextual distance may be calculated between a selected data point in a data sample and a data point in a contextual set of the selected data point. The contextual set may include the selected data point and one or more data points in the neighborhood of the selected data point. The contextual distance may be the difference between the selected data point's contribution to the integrity of the geometric structure of the contextual set and the data point's contribution to the integrity of the geometric structure of the contextual set. The process may be repeated for each data point in the contextual set of the selected data point. The process may be repeated for each selected data point in the data sample. A digraph may be created using a plurality of contextual distances generated by the process.
摘要:
Described is a technology, such as implemented in a computational software program, by which a minimal polynomial is efficiently determined for a radical expression over the ring Z of integer numbers or the ring Q of rational numbers. The levels of the radical are grouped into a level permutation group that is used to find a level permutation set. An annihilation polynomial is found based upon the level permutation set. The annihilation polynomial is factored, and a selection mechanism selects the minimal polynomial based upon the annihilation polynomial's factors.
摘要:
Methods and systems for converting text into natural personal handwriting are provided. One aspect relates to the training of a computer to recognize a user's handwriting style. In one embodiment, the computer receives handwriting samples of at least one character written by the user, such as the character being provided as the beginning, middle, or ending character among a plurality of other characters. Further embodiments allow for increased personalization of the handwriting. Another aspect relates to system and methods for displaying a representation of a computer user's handwriting. In one embodiment, the handwriting comprises variant shapes of letters, personalized connection style between letters, and connection parts that look pressure-sensitive. In another embodiment, characters are adjusted, such as cutting portions of the character to create a more realistic recreation and synthesis of the handwriting.
摘要:
Ink-parser-parameter optimization may be performed via parallel processing to accelerate searching for a set of optimal ink-parser parameters. Evaluators may parse pages of ink notes with different groups of parameters and may compute corresponding values for evaluation functions. Separate evaluation functions may be defined for the following types of ink-parker parsing engines: writing parser, writing/drawing classification, table detection, and list detection. A searcher may perform a grid-searching algorithm or a genetic algorithm to generate groups of parameters and may then pass the parameters to available evaluators for evaluation until evaluation-function values for a group of parameters satisfy a convergence condition.
摘要:
In accordance with embodiments of the invention, local metadata is embedded into an embedded interactive code document by combining a first m-array and a second m-array to generate a combined m-array with encoded local metadata such that a start position of the second m-array in the combined m-array is shifted, by an amount that is based on the local metadata, relative to a start position of the first m-array in the combined m-array. The first m-array and the second m-array may contain the same repeating bit sequence. Local metadata may be decoded from the embedded interactive code document by decoding the local metadata from the combined m-array by determining the amount by which the second m-array is shifted, relative to the first m-array, in the combined m-array.
摘要:
X-y positions of a digital pen's tip may be determined by using a calibration parameter to map the x-y positions of the respective centers of images captured by the pen's camera. The calibration parameter may be generated by iteratively calculating estimates of the calibration parameter. The iteratively calculated estimates may be based on a gradient of an error function of the iteratively generated estimates. Calibration input data may be produced by a user placing the pen tip in a fixed location on a surface, which may be a positionally encoded medium, such as paper, and then rotating the pen and/or moving the opposite end of the pen in various directions to capture multiple images for use in generating the calibration parameter. A user may perform such a calibration procedure without the need for complicated calibration equipment typically used in connection with conventional calibration techniques.
摘要:
Implementations of a revenue-based advertising auction are described. Bidders submit advertisements to be assigned to ad slots such that each advertisement has an associated bid vector that indicates a value known to the auctioneer that the bidder is offering to pay for placement of the advertisement in each of the ad slots, respectively. Advertisements are assigned to ad slots on a web page such that a community revenue is maximized given the available advertisements and bid vectors A price is calculated for each advertisement allocated to an ad slot based on a contribution to the community revenue attributed to the advertisement.
摘要:
Aspects of the present invention relate to the creation of an ink font. Based on characteristics of handwritten characters, the collection of characters may be scaled so as to adjust the size of the font to match predefined size values or relationships.