Google Inc.

Andrea Lynn Frome

Stephane Lafon

Andrew Timothy Szybalski

Stephen Chau

Derek King Prothro

Jerry Howard Morrison

Shlomo Reuben Urbach

Bo Wu

Huy M Nguyen

Luc Vincent

Alessandro Bissacco

Tal Yadid

Yuval Netzer

Noam Ben-Haim

Yoram Singer

Kong Man Cheung

Jeffrey Adgate Dean

Hartmut Neven

Hartwig Adam

Samy Bengio

Gregory Sean Corrado

Jonathon Shlens

Ahmad A Abdulkader

Raymond W Smith

Huy Tuan Nguyen

Marco Zennaro

Sergey Ioffe

Abhijit S Ogale

Charles Joseph Rosenberg

Erik Murphy-Chutorian

Radhika Malpani

Jingbin Wang

Dennis William Strelow

Tomas Mikolov

Sarah Moussa

German Cheung

Lance Williams

Mohammad Norouzi

Baris Yuksel

Ethan G Russell

Douglas Richard Grundman

Paul Eastlund

Dennis Hwang

Kansinee Adsanatham

Diego Ariel Gertzenstein

Howard Wellington Trickey

Melanie Clements

Alessandro Bissallo

Shlomo Urbach

Dennis Strelow

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for classifying data objects. One of the methods includes obtaining data that associates each term in a vocabulary of terms with a respective high-dimensional representation of the term; obtaining classification data for a data object, wherein the classification data includes a respective score for each of a plurality of categories, and wherein each of the categories is associated with a respective category label; computing an aggregate high-dimensional representation for the data object from high-dimensional representations for the category labels associated with the categories and the respective scores; identifying a first term in the vocabulary of terms having a high-dimensional representation that is closest to the aggregate high-dimensional representation; and selecting the first term as a category label for the data object.

Classifying data objects

Disclosed herein are methods and systems that relate to an 'activity assistant' that provides users with dynamically-selected 'activities' that are intelligently tailored to the user's world. The subject technology receives the one or more global parameters of one or more selected activities. The subject technology further receives the one or more account-specific parameters of a selected user account. For the selected user account, and for each of the one or more selected activities, the subject technology: (a) determines one or more signals based at least in part on one or more of the global parameters of the selected activity and one or more of the account-specific parameters of the selected user account, and (b) uses the determined signals as a basis for determining an importance of the selected activity for the selected user.

Activity assistant

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for classifying images. One of the methods includes obtaining data that associates each of a plurality of object category labels with a respective high-dimensional representation of the object category label, wherein the high-dimensional representation of the object category label is a numeric representation of the object category label in a high-dimensional space; receiving an input image; processing the input image using one or more core layers to generate an alternative representation of the input image; processing the alternative representation of the input image using a transformation layer to determine a high-dimensional representation for the input image; selecting, from the high-dimensional representations associated with the object category labels, a closest high-dimensional representation to the high-dimensional representation for the input image; and selecting the category label associated with the closest high-dimensional representation as a predicted label for the input image.

Classifying images

Some configurations of the subject technology provide a user interface including an activity-creation display configured to create a first activity of several activities, and a single-activity display configured to display a single activity of the several activities, and information related to a list of users associated with the activity. The single-activity display includes a parameter-entry display configured to select a parameter for the single activity and to send a signal to the single activity corresponding to the selected parameter, a heads-up display configured to send a heads-up message about the single activity to a user, a location display configured to display information related to a location of the single activity, a status indicator configured to display a status based on the status parameter of the single activity, and a status-entry display configured to, upon selection, send a signal to change the status indicator of the single activity.

User interface for activity status and history

A system and method is provided for automatically recognizing building numbers in street level images. In one aspect, a processor selects a street level image that is likely to be near an address of interest. The processor identifies those portions of the image that are visually similar to street numbers, and then extracts the numeric values of the characters displayed in such portions. If an extracted value corresponds with the building number of the address of interest such as being substantially equal to the address of interest, the extracted value and the image portion are displayed to a human operator. The human operator confirms, by looking at the image portion, whether the image portion appears to be a building number that matches the extracted value. If so, the processor stores a value that associates that building number with the street level image.

System and method of determining building numbers

Disclosed herein is an 'activity assistant' and an 'activity assistant user interface' that provides users with dynamically-selected &#x201c;activities&#x201d; that are intelligently tailored to the user's world. For example, a graphical UI includes selectable context elements, each of which corresponds to a user-attribute whose value provides a signal to the activity assistant. In response to selecting a parameter associated with at least one of the selectable context elements, a first signal is generated and provided to the activity assistant. In response to providing the signal, one or more activities are populated and ordered based, at least in part, on the signal, and subsequently displayed. The parameters may include a current mood of a user, a current location of the user, associations with other users, and a time during which the user desires to carry out the activity in some examples.

Providing user-defined parameters to an activity assistant

Systems and methods are provided for determining whether given business listings are spam or are closed businesses. In one example, a system receives a business listing from a user. The business listing contains a geolocation and the name of a business. The system also receives one or more street level images related to a geographic object that is at or near the same geolocation as the business listing. The system determines if the business listing is spam or is closed based on the received street level images by determining if the geographic object is the business or by analyzing whether the business associated with the listing is shown in any of the street level images.

Systems and methods of correlating business information to determine spam, closed businesses, and ranking signals

Configurations of an activity assistant processing system are disclosed that provides users with dynamically-selected activities that are intelligently tailored to a user. The subject technology accesses an index of activities. A score is then determined for each particular activity in the index of activities. The score is based on plurality of signals associated with user-specific parameters and global parameters to determine the quantitative importance of a particular activity. The subject technology identifies a subset of activities within the index of activities using the activity assistant processing system in which the subset includes at least one activity with a corresponding score that is greater than or equal to a decision threshold for the at least one activity. At least a portion of the subset of activities is then ranked based on the quantitative importance of the particular activity using the activity assistant processing system.

Determining activity importance to a user

Embodiments of this invention relate to detecting and blurring images. In an embodiment, a system detects objects in a photographic image. The system includes an object detector module configured to detect regions of the photographic image that include objects of a particular type at least based on the content of the photographic image. The system further includes a false positive detector module configured to determine whether each region detected by the object detector module includes an object of the particular type at least based on information about the context in which the photographic image was taken.

Object detection with false positive filtering

Systems and methods provide approximations of latitude and longitude coordinates of objects, for example a business, in street level images. The images may be collected by a camera. An image of a business is collected along with GPS coordinates and direction of the camera. Depth maps of the images may be generated, for example, based on laser depth detection or displacement of the business between two images caused by a change in the position of the camera. After identifying a business in one or more images, the distance from the camera to a point or area relative to the business in the one or more images may be determined based on the depth maps. Using this distance and the direction of the camera which collected the one or more images and GPS coordinates of the camera, the approximate GPS coordinates of the business may be determined.

System and method of determining map coordinates from images

Methods, systems and apparatus for identifying result images for a query image. One or more labels that are associated with the query image are obtained. Candidate images matching the query labels are identified. Visual similarity scores are generated for the candidate images. Each visual similarity score represents the visual similarity of a respective candidate image to the query image. Relevance scores are generated for each of the candidate images based on the visual similarity scores. Each relevance score represents a measure of relevance of the respective candidate images to the query image. The candidate images are ranked based on the relevance scores, a highest ranking subset of the candidate images being identified as result images and referenced by image search results. The result images can be candidate images that satisfy a similarity condition relative to the query image and other result images.

Query image search

Growing community of inventors

Berkeley, CA, United States of America

Andrea Lynn Frome