Dongmei Zhang

US Patent:

7400351, Jul 15, 2008

Filed:

Oct 6, 2004

Appl. No.:

10/959385

Inventors:

Dongmei Zhang - Redmond WA, US
Mehul Y Shah - Redmond WA, US

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

H04N 5/262

US Classification:

34824099, 348155

Abstract:

The subject invention provides a system and/or method that facilitates encoding a source image associated with a motion vector by employing at least one step image, wherein such step image provides a reduction in the amount of stored video. A step image component determines the maximum step image size and appropriate number of step images to simulate a smooth motion based upon an input component receiving data such as, but not limited to, a motion parameter and a computer environment parameter. Moreover, the step image component can utilize a motion controller component that facilitates the visual perception of smooth video during a zoom motion. Additionally, the step image component can utilize an encoder component providing specialized techniques in encoding with step images.

US Patent:

7528846, May 5, 2009

Filed:

Feb 23, 2005

Appl. No.:

11/064298

Inventors:

Dongmei Zhang - Bellevue WA, US
Mehul Y. Shah - Redmond WA, US
Vladimir Rovinsky - Redmond WA, US

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G09G 5/00
G06K 9/34
G06F 3/00
H04N 1/405

US Classification:

345619, 382173, 715700, 358 32

Abstract:

Systems and methods for adjusting a source image aspect ratio to match a different target display aspect ratio are described. In one aspect, the systems and methods determine that a source image has a first aspect ratio that is different than a second aspect ratio of a target display. Responsive to this determination, primary visual content of the source image is automatically detected. Dimensions and position of a crop of the source image are automatically calculated such that the crop preserves at least a subset of the primary visual content in the crop. Additionally, the crop is based on the second aspect ratio.

US Patent:

7805004, Sep 28, 2010

Filed:

Feb 28, 2007

Appl. No.:

11/680566

Inventors:

Shi Han - Beijing, CN
Yu Zou - Beijing, CN
Ming Chang - Beijing, CN
Peng Liu - Beijing, CN
Yi-Jian Wu - Beijing, CN
Lei Ma - Beijing, CN
Frank Soong - Beijing, CN
Dongmei Zhang - Bellevue WA, US
Jian Wang - Beijing, CN

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G06K 9/62

US Classification:

382185

Abstract:

Exemplary techniques are described for selecting radical sets for use in probabilistic East Asian character recognition algorithms. An exemplary technique includes applying a decomposition rule to each East Asian character of the set to generate a progressive splitting graph where the progressive splitting graph comprises radicals as nodes, formulating an optimization problem to find an optimal set of radicals to represent the set of East Asian characters using maximum likelihood and minimum description length and solving the optimization problem for the optimal set of radicals. Another exemplary technique includes selecting an optimal set of radicals by using a general function that characterizes a radical with respect to other East Asian characters and a complex function that characterizes complexity of a radical.

US Patent:

7890814, Feb 15, 2011

Filed:

Jun 27, 2007

Appl. No.:

11/823213

Inventors:

Dongmei Zhang - Redmond WA, US
Yingnong Dang - Beijing, CN
Xiaohui Hou - Beijing, CN
Song Huang - Beijing, CN
Jian Wang - Beijing, CN

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G06F 11/00

US Classification:

714 45, 714 48, 714 25, 717101

Abstract:

Described herein is technology for, among other things, accessing error report information. It involves various techniques and tools for analyzing and interrelating failure data contained in error reports and thereby facilitating developers to more easily and quickly solve programming bugs. Numerous parameters may also be specified for selecting and searching error reports. Several reliability metrics are provided to better track software reliability situations. The reliability metrics facilitate the tracking of the overall situation of failures that happen in the real word by providing metrics based on error reports (e. g. , failure occurrence trends, failure distributions across different languages).

US Patent:

7903877, Mar 8, 2011

Filed:

Mar 6, 2007

Appl. No.:

11/682722

Inventors:

Shi Han - Beijing, CN
Yu Zou - Beijing, CN
Ming Chang - Beijing, CN
Peng Liu - Beijing, CN
Yi-Jian Wu - Beijing, CN
Lei Ma - Beijing, CN
Frank Soong - Beijing, CN
Dongmei Zhang - Bellevue WA, US
Jian Wang - Beijing, CN

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G06K 9/00
G06K 9/18

US Classification:

382181, 382185

Abstract:

Exemplary methods, systems, and computer-readable media for developing, training and/or using models for online handwriting recognition of characters are described. An exemplary method for building a trainable radical-based HMM for use in character recognition includes defining radical nodes, where a radical node represents a structural element of an character, and defining connection nodes, where a connection node represents a spatial relationship between two or more radicals. Such a method may include determining a number of paths in the radical-based HMM using subsequence direction histogram vector (SDHV) clustering and determining a number of states in the radical-based HMM using curvature scale space-based (CSS) corner detection.

US Patent:

7953279, May 31, 2011

Filed:

Jun 28, 2007

Appl. No.:

11/823644

Inventors:

Xinjian Chen - Beijing, CN
Dongmei Zhang - Redmond WA, US
Yu Zou - Beijing, CN
Ming Chang - Beijing, CN
Shi Han - Beijing, CN
Jian Wang - Beijing, CN

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G06K 9/00
G06F 17/00

US Classification:

382187, 382155, 706 20, 715268

Abstract:

Described is a technology by which online recognition of handwritten input data is combined with offline recognition and processing to obtain a combined recognition result. In general, the combination improves overall recognition accuracy. In one aspect, online and offline recognition is separately performed to obtain online and offline character-level recognition scores for candidates (hypotheses). A statistical analysis-based combination algorithm, an AdaBoost algorithm, and/or a neural network-based combination may determine a combination function to combine the scores to produce a result set of one or more results. Online and offline radical-level recognition may be performed. For example, a HMM recognizer may generate online radical scores used to build a radical graph, which is then rescored using the offline radical recognition scores. Paths in the rescored graph are then searched to provide the combined recognition result, e. g. , corresponding to the path with the highest score.

US Patent:

7974472, Jul 5, 2011

Filed:

Jun 29, 2007

Appl. No.:

11/772032

Inventors:

Yu Zou - Beijing, CN
Ming Chang - Beijing, CN
Shi Han - Beijing, CN
Dongmei Zhang - Bellevue WA, US
Jian Wang - Beijing, CN

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G06K 9/00

US Classification:

382187

Abstract:

An exemplary method for online character recognition of East Asian characters includes acquiring time sequential, online ink data for a handwritten East Asian character, conditioning the ink data to produce conditioned ink data where the conditioned ink data includes information as to writing sequence of the handwritten East Asian character and extracting features from the conditioned ink data where the features include a tangent feature, a curvature feature, a local length feature, a connection point feature and an imaginary stroke feature. Such a method may determine neighborhoods for ink data and extract features for each neighborhood. An exemplary Hidden Markov Model based character recognition system may use various exemplary methods for training and character recognition.

US Patent:

7372536, May 13, 2008

Filed:

Mar 8, 2005

Appl. No.:

11/074414

Inventors:

Mehul Y. Shah - Redmond WA, US
Dongmei Zhang - Bellevue WA, US
Vladimir Rovinsky - Redmond WA, US

Assignee:

Microsoft Corporation - Redmond WA

International Classification:

G03B 21/32
G09G 5/00
G06T 13/00
G06K 9/36

US Classification:

352 87, 345473, 345619, 382236

Abstract:

A system and/or a method facilitates creating an image-based video by automatically applying a motion to an image. A motion component can receive the image and generate a corresponding motion, which can mitigate complications involved with authoring the image-based video. The motion component can apply the motion to the image based at least upon a comparison between an aspect ratio of the image and a resulting video aspect ratio. The comparison can be a numeric difference, wherein the difference can be associated with a category, wherein the category provides the motion to be applied.