Abstract: Systems, devices, and methods for side lobe control in holograms are described. The magnitude of the side lobes of a hologram depends on the distribution of refractive index modulation (?n), therefore control of side lobe magnitude may be achieved by controlling the distribution of ?n. The distribution of ?n may be controlled by replicating a hologram from a master with two reference beams, where the wavelength and angle of each reference beam, the playback angle of the master hologram, and the thickness of the master hologram, the copy holographic recording medium (HRM), and the recording substrate are carefully chosen to achieve a pattern of meta-interference within the HRM that matches the desired distribution of ?n.

Abstract: Methods, systems, and apparatus for receiving a query image and a user tap location, processing the received query image based on the user tap location, identifying one or more entities associated with the processed query image and in response to receiving (i) the query image, and (ii) the user tap location, providing information about the identified one or more of the entities.

Abstract: Methods, apparatus, systems, and computer-readable media are provided for tailoring composite graphical assistant interfaces for interacting with multiple different connected devices. The composite graphical assistant interfaces can be generated proactively and/or in response to a user providing a request for an automated assistant to cause a connected device to perform a particular function. In response to the automated assistant receiving the request, the automated assistant can identify other connected devices, and other functions capable of being performed by the other connected devices. The other functions can then be mapped to various graphical control elements in order to provide a composite graphical assistant interface from which the user can interact with different connected devices. Each graphical control element can be arranged to reflect how each connected device is operating simultaneous to the presentation of the composite graphical assistant interface.

Abstract: Implementations described herein relate to an automated assistant that iteratively renders various GUI elements as a user iteratively provides a spoken utterance, or sequence of spoken utterances, corresponding to a request directed to the automated assistant. These various GUI elements can be dynamically adapted as the user iteratively provides the spoken utterance to assist the user with efficiently completing the request. In some implementations, a generic container graphical element associated with candidate intent(s) can be initially rendered at a display interface of a computing device and dynamically adapted with tailored container graphical elements as a particular intent is determined while the user iteratively provides the spoken utterance.

Abstract: Techniques are disclosed for training and/or utilizing an automatic post-editing model in correcting translation error(s) introduced by a neural machine translation model. The automatic post-editing model can be trained using automatically generated training instances. A training instance is automatically generated by processing text in a first language using a neural machine translation model to generate text in a second language. The text in the second language is processed using a neural machine translation model to generate training text in the first language. A training instance can include the text in the first language as well as the training text in the first language.

Abstract: Systems and methods for data asset acquisition and obfuscation can be helpful for retrieving augmented reality rendering data assets from third parties. The sending of a software development kit and receiving back data assets can ensure the data assets are compatible with the augmented reality rendering experience in the user interface. The data acquisition system with obfuscation can also ensure the code generated by third parties is stripped of semantics and has reduced readability.

Abstract: A method includes receiving a training data set including a plurality of training data subsets. From two or more training data subsets in the training data set, the method includes selecting a support set of training examples and a query set of training examples. The method includes determining, using the classification model, a centroid value for each respective class. For each training example in the query set of training examples, the method includes generating, using the classification model, a query encoding, determining a class distance measure, determining a ground-truth distance, and updating parameters of the classification model. For each training example in the query set of training examples identified as being misclassified, the method further includes generating a standard deviation value, sampling a new query, and updating parameters of the confidence model based on the new query encoding.

Abstract: Systems and methods for snippet packet generation can include obtaining input data (e.g., input data descriptive of a gesture). The input data can be processed to determine a content item selected by the input. A snippet packet can be generated based on the content item, which can include the content item, address data, and location data. The snippet packet can be configured to be interacted with in order to navigate to the source web page of the content item including navigating to the specific portion of the web page that includes the content item.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training an image embedding model. In one aspect, a method comprises: obtaining training data comprising a plurality of training examples, wherein each training example comprises: an image pair comprising a first image and a second image; and selection data indicating one or more of: (i) a co-click rate of the image pair, and (ii) a similar-image click rate of the image pair; and using the training data to train an image embedding model having a plurality of image embedding model parameters.

Abstract: Implementations set forth herein relate an automated assistant that provides suggestion elements at an OEM application and/or third-party application that is accessible via an OEM vehicle computing device, which employs restrictions on communications between local applications. The automated assistant can render suggestions via OEM vehicle applications and/or third-party applications, and initialize actions for performance via the OEM vehicle applications and/or the third-party applications—despite the limitations. The suggestions can be rendered at a graphical user interface of a vehicle, and include content that, when spoken by a user, causes the automated assistant to initialize such actions. Actions that are suggested by the automated assistant can be performed by third-party applications that are different from a third-party application that is rendering the suggestion.

Abstract: The technology relates to integrating applications into a single user interface. One or more computing devices may display a user interface including associated data corresponding to a first application and generate a request for supplemental data. A second application which can satisfy the request for supplemental data may be identified by the one or more computing devices, and the one or more computing devices may generate and display a notification indicating that the second application can provide the requested supplemental data. A selection of the notification may be received by the one or more computing devices and a second user interface corresponding to the second application may be integrated into the user interface, wherein the second user interface displays the requested supplemental data.

Abstract: To provide content-aware audio navigation instructions, a client device executing a mapping application obtains one or more audio navigation directions for traversing from a starting location to a destination location along a route. The client device also identifies electronic media content playing from a source different from the mapping application which is executing at the client device or in proximity to the client device. The client device determines characteristics of the electronic media content and adjusts the audio navigation directions in accordance with the characteristics of the electronic media content. Then the client device presents the adjusted audio navigation directions to a user.

Abstract: The present disclosure is directed to a deep learning system for differential diagnoses of skin diseases. In particular, the system performs a method that can include obtaining a plurality of images that respectively depict a portion of a patient's skin. The method can include determining, using a machine-learned skin condition classification model, a plurality of embeddings respectively for the plurality of images. The method can include combining the plurality of embeddings to obtain a unified representation associated with the portion of the patient's skin. The method can include determining, using the machine-learned skin condition classification model, a skin condition classification for the portion of the patients skin, the skin condition classification produced by the machine-learned skin condition classification model by processing the unified representation, wherein the skin condition classification identifies one or more skin conditions selected from a plurality of potential skin conditions.

Abstract: In one general aspect, a method can include receiving, at a server computing device, a launch request to launch and stream media content pertaining to a game session, the launch request being provided by a mobile device, selecting, using the launch request, a media host configured to stream the media content, verifying a user associated with the mobile device and launching the game session for streaming to the user on a device other than the mobile device. In addition, the method can include during the game session and while streaming the media content, receiving a plurality of real time control requests from the mobile device, and executing the control request by compositing, in real time, the at least one change and the media content and to produce a composite display and transmitting, to the device other than the mobile device, the encoded composite display for streaming in real time.

Abstract: The techniques, methods, systems, and other mechanisms described herein include processes for determining if customized content should be generated, what information to include in the customized content, and when to provide the customized content. In general, a computing system determines that a user intends to travel to a physical venue. The computing system can determine if an entity associated with the physical venue has a web page. The computing system can determine various aspects of a predicted route of travel from the user's present location to the physical venue. The computing system can use location information indicating the user's current location and determine one or more routes of travel to the physical venue. The computing system can compare one or more determined aspects of the predicted route to threshold values to determine if customized content should be generated and presented to the user.

Abstract: Systems and methods for passing account authentication information via parameters. A server can provide, to a client device, an account parameter derived from an account credential used to authenticate a first application to insert into a link. The link can include an address referencing a second application. The account parameter can be passed from the first application to the second application responsive to an interaction on the link. The server can receive from the second application of the client device, subsequent to passing the account parameter from the first application to the second application, a request to authenticate the second application including the account parameter. The server can authenticate the client device for the second application using the account parameter. The server can transmit, responsive to authenticating the client device for the second application, an authentication indication to the second application of the client device.

Abstract: Conducting hands-free transactions comprises a server at a payment processing system, a user computing device, and a merchant computing device. The payment processing system receives a communication from a hands-free payment application on a user device, the communication comprising a first transaction token, an identification of a user account, and a beacon identifier. The merchant may provide a challenge to the user and use the response to identify the token and account of the user. The merchant computing device can use voice patterns of the user to assist in identifying the token and account of the user. The system receives from the merchant computing device a transaction request, the transaction request comprising the first transaction token and transaction data associated with the transaction request. The system determines that the transaction is for an amount less than a configured transaction limit and communicates a request for an authorization of the transaction.

Abstract: The disclosure is directed to embodiments for producing high-quality images simulating the application of a material (e.g., virtual cosmetics) to a person's body. Example implementations can generate an augmented image displaying a virtual cosmetic layer (e.g., lipstick) on a person's face. For instance, a computing system can obtain an image depicting the face and identify a region for applying the cosmetic. The system can use augmented reality and/or image filtering to process the image into datasets that can be combined with material data related to the virtual cosmetic to generate augmented image(s) simulating application of the material.

Abstract: A technique is provided to determine whether a region within a web page is viewable to a user through a browser window. Often, browsers will only show part of a web page at given time, creating a difficulty in establishing whether a region of the web page, which may be an advertisement for example, is in view. This is addressed by providing one or more test features within the region, rendering the web page, monitoring a behavioural characteristic of the test features and determining whether the region is in view based on the monitored behavioural characteristic, wherein the behavioural characteristic varies according to whether the test feature is currently being displayed. One example of a behavioural characteristic is a frame progression rate. Browsers will typically redraw elements of a web page at a higher rate if they are currently in view through the browser window, and this characteristic can therefore be used to determine whether the test feature, and thus the region, is in view.

Abstract: The present disclosure relates to a streaming speech-to-speech conversion model, where an encoder runs in real time while a user is speaking, then after the speaking stops, a decoder generates output audio in real time. A streaming-based approach produces an acceptable delay with minimal loss in conversion quality when compared to other non-streaming server-based models. A hybrid model approach for combines look-ahead in the encoder and a non-causal stacker with non-causal self-attention.