Abstract: A microelectronic chip device includes a semiconductor substrate and multiple on-chip strain sensors (OCSSs) constructed on the substrate at various locations of the substrate. The OCSSs may each include multiple piezoresistive devices configured to sense a strain at a location of the various locations and produce a strain signal representing the strain at that location. A strain measurement circuit may also be constructed on the semiconductor substrate and configured to measure strain parameters from the strain signals produced by the OCSSs. The strain parameters represent the strains at the various location. Values of the strain parameters can be used for analysis of mechanical stress on the chip device.

Abstract: A method includes obtaining a reference image and a target image each representing an environment containing moving features and static features. The method also includes determining an object mask configured to mask out the moving features and preserves the static features in the target image. The method additionally includes determining, based on motion parallax between the reference image and the target image, a static depth image representing depth values of the static features in the target image. The method further includes generating, by way of a machine learning model, a dynamic depth image representing depth values of both the static features and the moving features in the target image. The model is trained to generate the dynamic depth image by determining depth values of at least the moving features based on the target image, the object mask, and the static depth image.

Abstract: A switchable magnetic device comprising a first ferromagnetic material or layer comprising or consisting of a first rare earth nitride alloy, the first rare earth nitride alloy including at least two lanthanide species; a second ferromagnetic material or layer comprising or consisting of a second rare earth nitride alloy, the second rare earth nitride alloy including at least two lanthanide species; a blocking material or layer located between the first and second ferromagnetic materials or layers.

Abstract: A system can include an emitter structure connected to a reservoir containing a working material, wherein the working material is in electrical communication with a first electrode, an electrode opposing the emitter structure across a gap; and optionally a frame holding the emitter structure and the electrode.

Abstract: A set of user interface tools is described facilitating asynchronous adjudication of one or more regions-of-interest in a medical image by a group of two or more graders, each of which has access to the set of tools in a workstation environment. The set of tools include (1) a feature for enabling the graders to assess the medical image and manually delineate one or more specific regions-of-interest (ROI) in the medical image, (2) a feature for assessing the ROI(s) delineated by other graders, including display of the ROI delineated by other graders; (3) dialog features for explaining and discussing the assessments, including a text feature for discussing the assessments. The dialog features and the ROIs delineated by all the graders are visible to all the graders on the workstation display as they collectively adjudicate the medical image in a round-robin manner. The set of tools further include (4) a feature for manually verifying grader agreement with the other graders' assessments.

Abstract: Systems and methods are described herein for predicting tower downtimes in a cellular or wireless network and taking preemptive measures relative to the predicted downtime. One or more towers in a cellular network are selected to predict whether they are or are about to experience downtime. A prediction model trained to predict downtimes of towers sharing a characteristic with the one or more selected towers. Current information associated with the one or more selected towers is obtained and used in the prediction model to generate a downtime prediction for the one or more selected towers. The downtime prediction for the one or more selected towers is then provided to an administrator.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for audio-visual speech separation. A method includes: obtaining, for each frame in a stream of frames from a video in which faces of one or more speakers have been detected, a respective per-frame face embedding of the face of each speaker; processing, for each speaker, the per-frame face embeddings of the face of the speaker to generate visual features for the face of the speaker; obtaining a spectrogram of an audio soundtrack for the video; processing the spectrogram to generate an audio embedding for the audio soundtrack; combining the visual features for the one or more speakers and the audio embedding for the audio soundtrack to generate an audio-visual embedding for the video; determining a respective spectrogram mask for each of the one or more speakers; and determining a respective isolated speech spectrogram for each speaker.

Abstract: Systems and method are directed towards dynamically updating a remote management system of changes in e911 records. An e911 computing system may determine if an e911 record has been deleted from a local database. In response to determining that the e911 record has been deleted, the e911 computing system notifies a management system of the e911 record being deleted. The e911 computing system may also determine if the e911 record is new. In response to determining that the e911 record is new, the e911 computing system provides the e911 record to the management system. The e911 computing system may also determine if the e911 record has changed. In response to determining that the e911 record has changed: the e911 computing system generates a table of changes to the e911 record and provides the table of changes to the management system.

Abstract: Embodiments are directed towards providing document change alerts in a cloud-based wireless network system. A list of a plurality of target radio frequency design sheets is obtained. One or more cloud-based folders is then search for the plurality of target radio frequency design sheets. One or more corresponding target radio frequency design sheets are identified during the search. A last-modified date of the corresponding target radio frequency design sheet is compared to a target date. In response to the last-modified date of the corresponding target radio frequency design sheet being more recent than the target date: a most recent version of the corresponding target radio frequency design sheet is maintained and an older version of the corresponding target radio frequency design sheet is archived. A modification notification is then sent to at least one user associated with the corresponding target radio frequency design sheet that changed.

Abstract: Methods of modifying the composition of layers using selectively absorbing films are described. The composition of a layer can be modified by applying a selectively absorbing film in proximity to the applied coating and components of the layer can be selectively removed to provide a modified layers. The methods can be used to increase the concentration of particles in the layer.

Abstract: A method includes obtaining a reference image and a target image each representing an environment containing moving features and static features. The method also includes determining an object mask configured to mask out the moving features and preserves the static features in the target image. The method additionally includes determining, based on motion parallax between the reference image and the target image, a static depth image representing depth values of the static features in the target image. The method further includes generating, by way of a machine learning model, a dynamic depth image representing depth values of both the static features and the moving features in the target image. The model is trained to generate the dynamic depth image by determining depth values of at least the moving features based on the target image, the object mask, and the static depth image.

Abstract: A method includes obtaining a reference image and a target image each representing an environment containing moving features and static features. The method also includes determining an object mask configured to mask out the moving features and preserves the static features in the target image. The method additionally includes determining, based on motion parallax between the reference image and the target image, a static depth image representing depth values of the static features in the target image. The method further includes generating, by way of a machine learning model, a dynamic depth image representing depth values of both the static features and the moving features in the target image. The model is trained to generate the dynamic depth image by determining depth values of at least the moving features based on the target image, the object mask, and the static depth image.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for audio-visual speech separation. A method includes: obtaining, for each frame in a stream of frames from a video in which faces of one or more speakers have been detected, a respective per-frame face embedding of the face of each speaker; processing, for each speaker, the per-frame face embeddings of the face of the speaker to generate visual features for the face of the speaker; obtaining a spectrogram of an audio soundtrack for the video; processing the spectrogram to generate an audio embedding for the audio soundtrack; combining the visual features for the one or more speakers and the audio embedding for the audio soundtrack to generate an audio-visual embedding for the video; determining a respective spectrogram mask for each of the one or more speakers; and determining a respective isolated speech spectrogram for each speaker.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for audio-visual speech separation. A method includes: obtaining, for each frame in a stream of frames from a video in which faces of one or more speakers have been detected, a respective per-frame face embedding of the face of each speaker; processing, for each speaker, the per-frame face embeddings of the face of the speaker to generate visual features for the face of the speaker; obtaining a spectrogram of an audio soundtrack for the video; processing the spectrogram to generate an audio embedding for the audio soundtrack; combining the visual features for the one or more speakers and the audio embedding for the audio soundtrack to generate an audio-visual embedding for the video; determining a respective spectrogram mask for each of the one or more speakers; and determining a respective isolated speech spectrogram for each speaker.

Abstract: The present disclosure provides systems and methods to detect occluded objects using shadow information to anticipate moving obstacles that are occluded behind a corner or other obstacle. The system may perform a dynamic threshold analysis on enhanced images allowing the detection of even weakly visible shadows. The system may classify an image sequence as either “dynamic” or “static”, enabling an autonomous vehicle, or other moving platform, to react and respond to a moving, yet occluded object by slowing down or stopping.

Abstract: A method includes obtaining a reference image and a target image each representing an environment containing moving features and static features. The method also includes determining an object mask configured to mask out the moving features and preserves the static features in the target image. The method additionally includes determining, based on motion parallax between the reference image and the target image, a static depth image representing depth values of the static features in the target image. The method further includes generating, by way of a machine learning model, a dynamic depth image representing depth values of both the static features and the moving features in the target image. The model is trained to generate the dynamic depth image by determining depth values of at least the moving features based on the target image, the object mask, and the static depth image.

Abstract: A method includes obtaining a reference image and a target image each representing an environment containing moving features and static features. The method also includes determining an object mask configured to mask out the moving features and preserves the static features in the target image. The method additionally includes determining, based on motion parallax between the reference image and the target image, a static depth image representing depth values of the static features in the target image. The method further includes generating, by way of a machine learning model, a dynamic depth image representing depth values of both the static features and the moving features in the target image. The model is trained to generate the dynamic depth image by determining depth values of at least the moving features based on the target image, the object mask, and the static depth image.

Abstract: A minimally invasive controlled drug delivery system for delivering a particular drug or drugs to a particular location of the eye, the system including a porous film template having pores configured and dimensioned to at least partially receive at least one drug therein, and wherein the template is dimensioned to be delivered into or onto the eye.

Abstract: A minimally invasive controlled drug delivery system for delivering a particular drug or drugs to a particular location of the eye, the system including a porous film template having pores configured and dimensioned to at least partially receive at least one drug therein, and wherein the template is dimensioned to be delivered into or onto the eye.

Abstract: Quantum dot-based color converters having a high density of sub-pixels. The sub-pixels have a high density of quantum dots that provide for high conversion efficiency within small sub-pixel aspect ratios and small volumes. The color converter can be used in optical displays.