Microsoft Technology Licensing, LLC

Broadcom Corporation

Other

Simon John Baker

Anitha Kannan

Krishnan Ramnath

Dahua Lin

Richard S Szeliski

Gang Hua

Ashish Kapoor

Qifa Ke

Li Fung Chang

Yongjun Wu

Sing Bing Kang

Mark Hahm

Charles Lawrence Zitnick, Iii

Sudipta Narayan Sinha

Matthew Richard Wozniak

Eitan Sharon

Eran Borenstein

Mehmet Nejat Tek

Kevin A Chin

Padmananbhan Anandan

Catalin Alexandru Negrila

Venkata S K Kamal Lanka

Brian Kohlwey

Edward Hsiao

Bin Liu

Matthew Uyttendaele

Michael Isard

Shengyang Dai

Eric Paul Bennett

Jia H Deng

Jiawei Huang

Joshua Phillips

Xintian Li

Amir Akbarzadeh

Dexin Li

Charles Zitnick

Juliet Anne Bernstein

Gerhard Florian Schroff

Khoi Ly

Images stored in an information repository are prepared for browsing. For each image in the repository, text in the repository is mined to extract snippets of text about the image which are semantically relevant to the image, and for each of these snippets of text, keyterms are detected in the snippet of text which represent either concepts that are related to the image or entities that are related to the image, and the snippet of text and keyterms are associated with the image. Each keyterm that is associated with each image in the repository is hyperlinked to each other image in the repository that has this keyterm associated therewith. A graphical user interface allows a user to browse the images in the repository by using their associated snippets of text and hyperlinked keyterms.

Browsing images via mined hyperlinked text snippets

A method described herein includes receiving a digital image, wherein the digital image includes a first element that corresponds to a first domain and a second element that corresponds to a second domain. The method also includes automatically assigning a label to the first element in the digital image based at least in part upon a computed probability that the label corresponds to the first element, wherein the probability is computed through utilization of a first model that is configured to infer labels for elements in the first domain and a second model that is configured to infer labels for elements in the second domain. The first model receives data that identifies learned relationships between elements in the first domain and elements in the second domain, and the probability is computed by the first model based at least in part upon the learned relationships.

Image tagging based upon cross domain context

The present invention discloses fiducial marker systems or tag systems and methods to detect and decode a tag. The tag comprises a plurality of cells arranged in a grid pattern. The plurality of cells forms a square or rectangular shape. Each cell is in square or rectangular shape and has either black, white, or gray grayscale. Any two cells that share a common borderline should have different grayscales; and the four corner cells of the tag are all in black. The method for detecting and decoding such a tag comprises detecting a plurality of connected cells in an image, determining whether the plurality of connected cells in the image have all three grayscales, determining whether total number of cells in the plurality of connected cells is greater than a predetermined threshold, determining whether any two cells in the plurality of connected cells that share a common borderline in the image have different grayscales, and decoding the tag based on the grayscales of the plurality of connected cells in the image.

Fiducial marker based on blocks of different grayscales

A system and method are disclosed for quickly and easily browsing to points of interest within a recorded video. In examples, the present system works in tandem with a segment list which includes a list of different video sequences of a stored video of an event. One typical example of a segment list is a play-by-play (PBP) which is prepared contemporaneously with a sporting event. Segments from a segment list may be associated with, or indexed to, corresponding sequences from the stored video. Thereafter, a user interface allows a user to interactively watch or browse the event by jumping to a desired point in the video for playback by selecting its associated segment from the user interface.

Browsing videos via a segment list

The presentation style of a video is identified. A set of features that represents the video is computed. A pre-learned video presentation style classifier is then used to weight each of the features in the set of features and determine a presentation style that is predominately employed in the video based on the weighting of the features.

Identifying presentation styles of educational videos

In one embodiment, a video processing systemmay filter a video data set to correct skew and wobble using a central processing unitand a graphical processing unit. The video processing systemmay apply a rolling shutter effect correction filter to an initial version of a video data set. The video processing systemmay simultaneously apply a video stabilization filter to the initial version to produce a final version video data set.

Joint video stabilization and rolling shutter correction on a generic platform

The claimed subject matter provides for systems and/or methods for identification of instances of an object of interest in 2D images by creating a database of 3D curve models of each desired instance and comparing an image of an object of interest against such 3D curve models of instances. The present application describes identifying and verifying the make and model of a car from a possibly single image&#x2014;after the models have been populated with training data of test images of many makes and models of cars. In one embodiment, an identification system may be constructed by generating a 3D curve model by back-projecting edge points onto a visual hull reconstruction from silhouettes of an instance. The system and methods employ chamfer distance and orientation distance provides reasonable verification performance, as well as an appearance model for the taillights of the car to increase the robustness of the system.

Object identification using 3-D curve matching

The description relates to sentence autocaptioning of images. One example can include a set of information modules and a set of sentence generation modules. The set of information modules can include individual information modules configured to operate on an image or metadata associated with the image to produce image information. The set of sentence generation modules can include individual sentence generation modules configured to operate on the image information to produce a sentence caption for the image.

Autocaptioning of images

Image search results are obtained by providing weights to visual features to emphasize features corresponding to objects of interest while simultaneously deemphasizing irrelevant or inconsistent features that lead to poor search results. In order to minimize the impact of visual features that are unreliable or irrelevant with respect to the objects of interest in the image, context-dependent weights are provided to detect visual features such that those visual features pertaining to the objects of interest are more heavily weighted than those visual features that pertain to irrelevant or unreliable portions of the image. Visual features may be weighted for images in a searchable database. Training data may be obtained and used in weighting visual features in a query image and, alternatively, in searchable database images.

Image retrieval using discriminative visual features

Text in web pages or other text documents may be classified based on the images or other objects within the webpage. A system for identifying and classifying text related to an object may identify one or more web pages containing the image or similar images, determine topics from the text of the document, and develop a set of training phrases for a classifier. The classifier may be trained and then used to analyze the text in the documents. The training set may include both positive examples and negative examples of text taken from the set of documents. A positive example may include captions or other elements directly associated with the object, while negative examples may include text taken from the documents, but from a large distance from the object. In some cases, the system may iterate on the classification process to refine the results.

Matching text to images

Growing community of inventors

Palo Alto, CA, United States of America

Simon John Baker