Abstract: Embodiments relate to extracting features from images, such as by identifying keypoints and generating keypoint descriptors of the keypoints. An apparatus includes a pyramid image generator circuit, a keypoint descriptor generator circuit, and a pyramid image buffer. The pyramid image generator circuit generates an image pyramid from an input image. The keypoint descriptor generator circuit processes the pyramid images for keypoint descriptor generation. The pyramid image buffer stores different portions of the pyramid images generated by the pyramid image generator circuit at different times and provides the stored portions of the pyramid images to the keypoint descriptor generator circuit for keypoint descriptor generation. When first portions of the pyramid images are no longer needed for the keypoint descriptor generation, the first portions are removed from the pyramid image buffer to provide space for second portions of the pyramid images that are needed for the keypoint descriptor generation.

Abstract: A configuration for a breathing sequence may be defined using a user interface of a user device. The user interface may also be used to begin the breathing sequence. During the breathing sequence, a fluctuating user interface element may fluctuate at a cyclic rate. Such fluctuation may include repeated growing and repeated shrinking of the fluctuating user interface element. During the breathing sequence, heart rate data may be collected and used to present heart rate information at a conclusion of the breathing sequence.

Abstract: An electronic device can include a housing component at least partially defining an internal volume and including a wall. A component can be disposed within the wall and at least partially surrounded by a material forming the housing. A battery can be disposed in the internal volume and electrically connected to the component.

Abstract: The subject technology provides a framework for learning neural scene representations directly from images, without three-dimensional (3D) supervision, by a machine-learning model. In the disclosed systems and methods, 3D structure can be imposed by ensuring that the learned representation transforms like a real 3D scene. For example, a loss function can be provided which enforces equivariance of the scene representation with respect to 3D rotations. Because naive tensor rotations may not be used to define models that are equivariant with respect to 3D rotations, a new operation called an invertible shear rotation is disclosed, which has the desired equivariance property. In some implementations, the model can be used to generate a 3D representation, such as mesh, of an object from an image of the object.

Abstract: An electronic device displays a messaging user interface of a message application, including a conversation transcript of a messaging session between a user of the electronic device and at least one other user, and a first message region that includes a current version of a first message in the messaging session. The conversation transcript includes an indication that the current version of the first message is a revised version of the first message in the conversation transcript. In response to detecting an input that corresponds to a request to see versions of the first message, the electronic device displays a versions user interface that includes the current version of the first message and one or more prior versions of the first message.

Abstract: Rendering an avatar in a selected environment may include determining as inputs into an inferred shading network, an expression geometry to be represented by an avatar, head pose, and camera angle, along with a lighting representation for the selected environment. The inferred shading network may then generate a texture of a face to be utilized in rendering the avatar. The lighting representation may be obtained as lighting latent variables which are obtained from an environment autoencoder trained on environment images with various lighting conditions.

Abstract: The present disclosure can provide devices, methods, and computer-readable medium for secure frame management. The techniques disclosed herein provide detection of triggering items in one or more frames of streaming video. The camera transmits one or more frames of the video over a network to a computing device. The computing device can begin a streaming session with a server and stream the one or more frames of video and accompanying metadata to the server. The frames, metadata, and associated keys can all be encrypted prior to streaming to the server. For each subsequent segment of video frames that includes the triggering item, the server can append the frames of that segment to the video clip in an encrypted container.

Abstract: In some implementations, a user device can coordinate adjustments to a composite GUI generated in cooperation with an accessory device that presents the composite graphical user interface. For example, the user device can receive configuration data that defines the maximum GUI display area (e.g., size, dimensions) of the accessory device, various view areas within the GUI display area of the accessory device into which the user device can render GUIs, and/or transitions between the various view areas. The user device can generate a display buffer based on the maximum GUI display area, generate a graphical user interface in an area of the buffer corresponding to a current view area specified by the accessory device, and send to the accessory device video frames corresponding to the display buffer. The accessory device can generate a composite GUI based on the received video frames that include the user device generated GUI.

Abstract: Techniques described herein include receiving selection of information identifying one or more members for participating in a fitness session and a session type of the fitness session. The techniques also includes determining fitness related notification transmit parameters for a first member of the one or more members. The techniques also includes presenting one or more notifications from a second computing device associated with a second member of the one or more members at a rate, wherein the rate is based at least in part on the session type.

Abstract: A multi-layer low-pass filter is used to filter a first frame of video data representing at least a portion of an environment of an individual. A first layer of the filter has a first filtering resolution setting for a first subset of the first frame, while a second layer of the filter has a second filtering resolution setting for a second subset. The first subset includes a data element positioned along a direction of a gaze of the individual, and the second subset of the frame surrounds the first subset. A result of the filtering is compressed and transmitted via a network to a video processing engine configured to generate a modified visual representation of the environment.

Abstract: Systems, methods, and apparatuses disclosed herein can compensate for imbalances between multiple transport streams of an uplink (UL)-multiple-input and multiple-output (MIMO) transmission. These systems, methods, and apparatuses can adjust one or more signal metrics of multiple recovered transport streams of the UL-MIMO transmission that are received in the presence of imbalances between the multiple transport streams of the UL-MIMO transmission. As an example, these systems, methods, and apparatuses can inject noise into one or more of the multiple recovered transport streams to adjust the one or more signal metrics of the multiple recovered transmission streams to lessen the imbalances between the multiple transmission streams.

Abstract: A foldable electronic device may have a foldable housing. The foldable housing may be configured to bend about a bend axis. First and second portions of the housing that rotate relative to each other may be coupled by a hinge that is aligned with the bend axis. A foldable display may be coupled to the foldable housing and may be configured to bend along the bend axis as the foldable housing is folded. The display may have an array of pixels supported by a metal layer. The pixels may be interposed between a display cover layer and the metal layer. The foldable housing may have a support layer. To help support the display for bending about the bend axis while preventing damage to the display when the display is contacted by an external object, a spring layer may be interposed between the metal layer and the support layer.

Abstract: A first cellular base station transmits a configuration message to a reporting device installed on a high-speed vehicle. The configuration message specifies one or more parameters of a Doppler measurement report. The reporting device performs one or more first Doppler measurements on the first base station and/or one or more second Doppler measurements on a second base station. The reporting device transmits the Doppler measurement report to the first and/or second base stations. The Doppler measurement report may be used by the first and/or second base stations to perform Doppler pre-compensation on transmissions to the reporting device.

Abstract: A switchable panel is described that includes exterior layers, a switchable component disposed between the exterior layers, a lamination frame extending from outer edges of the switchable component to outer edges of the panel between the exterior layers, and spacers embedded within the lamination frame. The spacers are configured to prohibit the exterior layers from moving toward each other at the outer edges of the panel when a central portion of the panel experiences tension.

Abstract: The present disclosure relates to determining when the head position of a user viewing user interfaces in a computer-generated reality environment is not in a comfortable and/or ergonomic position and repositioning the displayed user interface so that the user will reposition her/his head to view the user interface at a more comfortable and/or ergonomic head position.