Google Inc.

Sony Corporation

Microsoft Technology Licensing, LLC

Omron Corporation

Northwestern University

Nec Laboratories America, Inc.

Shengyang Dai

Su Wang

Akira Nakamura

Erik Murphy-Chutorian

Jun Yokono

Mei Han

Takeshi Ohashi

Vivek Kwatra

Xun Xu

Shinko Yuanhsien Cheng

Jingyu Cui

Matthew Stephen Steiner

David Cohen

Aravind Krishnaswamy

Jiang Wang

Eric Charles Henry

Ivan Neulander

Flavio Lerda

Timothy Sepkoski St Clair

Jay Steele

Qi Gu

Pardis Beikzadeh

Koji Ashida

Matthias Grundmann

David Foster Lieb

John Patrick Flynn

Yang Xu

Teresa Ko

Kyle Howard Heath

Troy Chinen

Madhur Khandelwal

James A Gallagher

Nicholas Butko

Jason Chang

Ehud Rivlin

Kedar Kanitkar

Loren Puchalla Fiore

Kristina Tiffany Eng

Florian Kriener

Eunyoung Kim

Nan Wang

Vahid Kazemi

John Spiegel

Falk Sticken

Tom Binder

David Loxton

Steven Vandebogart

Nikolaos Trogkanis

Ingo Wehmeyer

Ying Wu

Sing Bing Kang

Sergey Ioffe

Charles Joseph Rosenberg

Ming-Hsuan Yang

Simon John Baker

Yihong Gong

Jia Li

Shihong Lao

Wei Xu

Huizhong Chen

Xiangxin Zhu

Ming Yang

Joshua Foster Slocum

Jilin Tu

Jiyang Gao

Chuck Joseph Rosenberg

Xutao Lv

Xiaojie Jin

A computing system for detecting objects in an image can perform operations including generating an image pyramid that includes a first level corresponding with the image at a first resolution and a second level corresponding with the image at a second resolution. The operations can include tiling the first level and the second level by dividing the first level into a first plurality of tiles and the second level into a second plurality of tiles; inputting the first plurality of tiles and the second plurality of tiles into a machine-learned object detection model; receiving, as an output of the machine-learned object detection model, object detection data that includes bounding boxes respectively defined with respect to individual ones of the first plurality of tiles and the second plurality of tiles; and generating image object detection output by mapping the object detection data onto an image space of the image.

Systems and methods for object detection using image tiling

The present disclosure provides systems and methods that include or otherwise leverage an object detection training model for training a machine-learned object detection model. In particular, the training model can obtain first training data and train the machine-learned object detection model using the first training data. The training model can obtain second training data and input the second training data into the machine-learned object detection model, and receive as an output of the machine-learned object detection model, data that describes the location of a detected object of a target category within images from the second training data. The training model can determine mined training data based on the output of the machine-learned object detection model, and train the machine-learned object detection model based on the mined training data.

Noise tolerant ensemble RCNN for semi-supervised object detection

Some implementations related to archiving of functional images. In some implementations, a method includes accessing images and determining one or more functional labels corresponding to each of the images and one or more confidence scores corresponding to the functional labels. A functional image score is determined for each of the images based on the functional labels having a corresponding confidence score that meets a respective threshold for the functional labels. In response to determining that the functional image score meets a functional image score threshold, a functional image signal is provided that indicates that one or more of the images that meet the functional image score threshold are functional images. The functional images are determined to be archived, and are archived by associating an archive attribute with the functional images such that functional images having the archive attribute are excluded from display in views of the images.

Functional image archiving

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for converting unstructured documents to structured key-value pairs. In one aspect, a method includes: providing an image of a document to a detection model, wherein: the detection model is configured to process the image to generate an output that defines one or more bounding boxes generated for the image; and each bounding box generated for the image is predicted to enclose a key-value pair including key textual data and value textual data, wherein the key textual data defines a label that characterizes the value textual data; and for each of the one or more bounding boxes generated for the image: identifying textual data enclosed by the bounding box using an optical character recognition technique; and determining whether the textual data enclosed by the bounding box defines a key-value pair.

Identifying key-value pairs in documents

Methods, and systems, including computer programs encoded on computer storage media for neural network architecture search. A method includes defining a neural network computational cell, the computational cell including a directed graph of nodes representing respective neural network latent representations and edges representing respective operations that transform a respective neural network latent representation; replacing each operation that transforms a respective neural network latent representation with a respective linear combination of candidate operations, where each candidate operation in a respective linear combination has a respective mixing weight that is parameterized by one or more computational cell hyper parameters; iteratively adjusting values of the computational cell hyper parameters and weights to optimize a validation loss function subject to computational resource constraints; and generating a neural network for performing a machine learning task using the defined computational cell and the adjusted values of the computational cell hyper parameters and weights.

Resource constrained neural network architecture search

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for converting unstructured documents to structured key-value pairs. In one aspect, a method comprises: providing an image of a document to a detection model, wherein: the detection model is configured to process the image to generate an output that defines one or more bounding boxes generated for the image; and each bounding box generated for the image is predicted to enclose a key-value pair comprising key textual data and value textual data, wherein the key textual data defines a label that characterizes the value textual data; and for each of the one or more bounding boxes generated for the image: identifying textual data enclosed by the bounding box using an optical character recognition technique; and determining whether the textual data enclosed by the bounding box defines a key-value pair.

Implementations relate to sharing images and image albums over a communication network. In some implementations, a computer-implemented method includes determining that a shared album data structure is accessible by a device and includes references to album images. The device determines one or more suggested images from a collection of stored images associated with a user, based on determining sharing scores for the stored images. The sharing scores are based on comparing one or more characteristics of the stored images to one or more corresponding characteristics of the album images. The method includes causing display of the one or more suggested images by the device, receiving a selection of at least one selected image of the suggested images based on received user input, and causing an update of the shared album data structure with the at least one selected image to be accessible over a communication network by recipient users.

Sharing images and image albums over a communication network

A method includes grouping media items associated with a user into segments based on a timestamp associated with each media item and a total number of media items. The method also includes selecting target media from the media items for each of the segments based on media attributes associated with the media item. The method also includes generating a video that includes the target media for each of the segments by generating a first animation that illustrates a first transition from a first item from the target media to a second item from the target media with movement of the first item from an onscreen location to an offscreen location, wherein the first item is adjacent to the second item in the first animation and determining whether the target media includes one or more additional items. The method also includes adding a song to the video.

Generating videos of media items associated with a user

Implementations generally relate to generating image compositions. In some implementations, a method includes receiving a plurality of photos from a user and determining one or more composition types from the photos. The method further includes generating one or more compositions from the received photos based on the one or more determined composition types, where each composition is based on modified foregrounds of the photos. The method further includes providing the one or more generated compositions to the user.

Generating image compositions

Implementations generally relate to generating compositional media content. In some implementations, a method includes receiving a plurality of photos from a user, and determining one or more composition types from the photos. The method also includes generating compositions from the selected photos based on the one or more determined composition types. The method also includes providing the one or more generated compositions to the user.

Generating compositions

Content-based retrieval of digital pathology images (DPI) is a fundamental component in an intelligent DPI processing and management system. One key issue of content-based DPI retrieval is how to represent an image as a feature vector, capturing its key information that is most relevant to the goal of retrieval. A unified framework of extracting information of different levels for DPI, namely low level color information, middle level texture information and high level diagnostic information is described herein. Such information from all the levels are integrated to the end of content-based DPI retrieval.

Combining information of different levels for content-based retrieval of digital pathology images

Growing community of inventors

San Jose, CA, United States of America

Shengyang Dai