Abstract: Apparatuses, systems, and techniques are presented to select neural networks. In at least one embodiment, one or more first neural networks can be used to select one or more second neural networks, as may be based at least in part upon an inference to be generated by the one or more second neural networks.

Abstract: Apparatuses, systems, and techniques are presented to classify objects in images. In at least one embodiment, one or more neural networks are used to identify one or more objects in one or more full images based, at least in part, on the one or more neural networks having been trained using the one or more full images and one or more portions of the one or more full images.

Abstract: Aspects of the present disclosure are directed to a multi-sensor don/doff detection system for an artificial reality device headset. The multi-sensor don/doff detection system can use a combination of a proximity sensor, an inertial measurement unit (IMU), and an eye tracking/face tracking (ET/FT) unit to make these determinations. However, when both the ET/FT system and proximity sensor system are active, they can have system coexistence issues. Thus, only one of these systems can be used simultaneously. The multi-sensor don/doff detection system can more accurately identify don events by using input from the proximity sensor and the IMU. The multi-sensor don/doff detection system can also more accurately identify doff events by using input from the IMU and either A) the proximity sensor when the ET/FT system is disabled or B) the ET/FT system when the ET/FT system is enabled.

Abstract: A vehicle lamp assembly, a vehicle lamp and a corresponding motor vehicle. The vehicle lamp assembly including a support, for supporting a light emitting body. The vehicle lamp assembly further includes a housing provided to overlap with at least a part of the support and a first gap formed between the overlapping portions of the housing and the support. The vehicle lamp assembly additionally includes a first light blocking layer, with the first light blocking layer being provided between the overlapping portions of the housing and the support, and at least partially filling the first gap, so as to prevent light from the light emitting body from emitting outwards through the first gap.

Abstract: Apparatuses, systems, and techniques are presented to predict annotations for objects in images. In at least one embodiment, boundaries of an object within an image can be identified based, at least in part, on a user-generated outline of only a portion of this object or information about a size of this object provided by a user.

Abstract: Apparatuses, systems, and techniques are presented to predict annotations for objects in images. In at least one embodiment, one or more neural networks are used to help generate one or more segmentation boundaries of one or more objects within one or more digital images, wherein the one or more neural networks are to transform one or more representations of one or more portions of the one or more objects into one or more lower-dimensional representations of the one or more portions of the one or more objects.

Abstract: A power tool system includes a tool body for implementing the function of the power tool and a battery pack detachably connectable to the tool body. The battery pack also includes a housing and a cell assembly disposed in the housing and used for storing or releasing electrical energy. The battery pack also includes a battery management module electrically connected to a detection module and the cell assembly and configured to dynamically adjust a charge parameter of the battery pack or a discharge parameter of the battery pack according to the current cycle count of the battery pack.

Abstract: Apparatuses, systems, and techniques to generate one or more images of an object. In at least one embodiment, a technique includes training one or more neural networks to generate one or more images of an object from at least a first image of the object and a second lower-resolution image of the object, where the training includes a comparison of the one or more generated images of the object with the second lower-resolution image of the object.

Abstract: Apparatuses, systems, and techniques to indicate an extent, to which text corresponds to one or more images. In at least one embodiment, an extent to which text corresponds to one or more images is indicated using one or more neural networks and used to train the one or more neural networks.

Abstract: Examples disclosed herein relate to an apparatus and method of forming thin film layers on a substrate. A first piezoelectric material layer is deposited on the substrate in a first chamber. The first piezoelectric material layer is formed on the substrate while the substrate is at a first temperature. A second piezoelectric material layer is deposited on the first piezoelectric material layer after cooling the substrate to a second temperature. The second temperature is lower than the first temperature. The first piezoelectric material layer and the second piezoelectric material layer both comprise a first piezoelectric material.

Abstract: Electrically conductive masking tapes include an electrically conductive backing and an electrically conductive pressure sensitive adhesive layer. The pressure sensitive adhesive contains an acrylate-based copolymeric matrix, a crosslinker, an electrically conductive filler, and at least one antioxidant. The acrylate-based copolymeric matrix is the reaction product of a polymerizable mixture including at least one first alkyl(meth)acrylate monomer with a homopolymer Tg of less than ?50° C., and at least one hydroxyl-functional alkyl(meth)acrylate with a homopolymer Tg of less than ?10° C. The electrically conductive tape is capable of being laminated to and cleanly removed from a substrate surface, after being subjected to harsh conditions such as plasma vapor deposition conditions.

Abstract: An obstacle recognition method and apparatus, and a related device are provided. The method includes: obtaining an actual path loss value between a first AP and a second AP, path loss value pairs between a terminal and a plurality of AP pairs, and path loss values of the plurality of AP pairs; obtaining, based on the path loss value pairs between the terminal and the plurality of AP pairs, an AP pair similar to an AP pair formed by the first AP and the second AP; obtaining a second path loss value between the first AP and the second AP based on a path loss value of the similar AP pair; and comparing the second path loss value between the first AP and the second AP with the actual path loss value, to determine whether an obstacle exists between the first AP and the second AP.

Abstract: The present invention provides scanning mechanisms for imaging probes using for imaging mammalian tissues and structures using high resolution imaging, including high frequency ultrasound and/or optical coherence tomography. The imaging probes include adjustable rotational drive mechanism for imparting rotational motion to an imaging assembly containing either optical or ultrasound transducers which emit energy into the surrounding area. The imaging assembly includes a scanning mechanism having including a movable member configured to deliver the energy beam along a path out of said elongate hollow shaft at a variable angle with respect to said longitudinal axis to give forward and side viewing capability of the imaging assembly. The movable member is mounted in such a way that the variable angle is a function of the angular velocity of the imaging assembly.

Abstract: Aspects of the disclosure provide a sensing apparatus including a sensing device, a memory, and processing circuitry. The sensing device includes resonators having respective resonant frequencies. The resonators include an array of inductive coils positioned on a surface of the apparatus. The sensing device can output a signal indicating changes of the resonant frequencies caused by presence of an object proximate to the surface. The memory stores reference signals corresponding to reference objects. Each reference signal indicates changes of the resonant frequencies caused by the respective reference object proximate to the surface. The processing circuitry can receive, from the sensing device, a particular signal indicating changes of the resonant frequencies caused by presence of a particular object proximate to the surface. The processing circuitry compares the particular signal with the stored reference signals of the reference objects to determine an identity of the particular object.

Abstract: A battery pack includes a housing of a first material, a cell assembly, and a cell support of a second material. The cell assembly is disposed in the housing and includes a plurality of cell units. The cell unit includes a positive electrode of the cell unit and a negative electrode of the cell unit. The cell support is configured to support at least the cell assembly. The cell support is at least disposed at two ends of the cell assembly and at least part of the cell support encapsulates the positive electrode of the cell unit and the negative electrode of the cell unit. The first material is different from the second material.

Abstract: Apparatuses, systems, and techniques to improve federated learning for neural networks. In at least one embodiment, a federated server dynamically selects neural network weights according to one or more learnable aggregation weights indicating a contribution from each of one or more edge devices or clients during federated training according to various characteristics of each edge device or client model and training data.

Abstract: The present invention relates to novel antibodies, particularly antibodies directed against deletion mutants of epidermal growth factor receptor and particularly to the type III deletion mutant, EGFRvIII. The invention also relates to human monoclonal antibodies directed against deletion mutants of epidermal growth factor receptor and particularly to EGFRvIII. Diagnostic and therapeutic formulations of such antibodies, and immunoconjugates thereof, are also provided.

Abstract: A battery pack includes a housing of a first material, a cell assembly, and a cell support of a second material. The cell assembly is disposed in the housing and includes a plurality of cell units. The cell unit includes a positive electrode of the cell unit and a negative electrode of the cell unit. The cell support is configured to support at least the cell assembly. The cell support is at least disposed at two ends of the cell assembly and at least part of the cell support encapsulates the positive electrode of the cell unit and the negative electrode of the cell unit. The first material is different from the second material.

Abstract: The present invention relates to a WPT coil and an apparatus having same, and more particularly, to a WPT coil configured to have a three-dimensional shape by using a flexible substrate or the like, and an apparatus having same. A WPT coil is provided in a WPT device to transmit or receive wireless power, according to the present invention, the WPT coil comprising: a flexible substrate; and a coil structure provided on the flexible substrate, wherein the WPT coil forms a three-dimensional shape rather than a flat plate shape.

Abstract: Apparatuses, systems, and techniques to select a neural network using an amount of memory to be used. In at least one embodiment, a processor includes one or more circuits to cause one or more neural networks to be selected from a plurality of neural networks based, at least in part, on an amount of memory to be used by the one oe more neural networks.