Abstract: Introduced here is an acoustic collection device that is designed to extend the path along which sound waves can be collected by a microphone. An acoustic collection device can include a tubular body that has (i) a distal interface through which sound waves corresponding to sounds internal to a living body are collected and (ii) a proximal interface through which the sound waves are presented to a microphone. A channel defined by the inner surface of the tubular body extends between the distal and proximal interfaces. The channel allows the sound waves to travel from the distal interface to the proximal interface. The acoustic collection device could be connected to the housing of an electronic device that is used to record the sound waves, or the acoustic collection device could be implemented in an input unit for an electronic stethoscope system.

Abstract: Introduced here are electronic stethoscope systems designed to simultaneously monitor sounds originating from within a body under examination and the ambient environment. An electronic stethoscope system can include one or more input units that are connected to a hub unit. Each input unit may have at least one auscultation microphone and at least one ambient microphone. To improve the quality of sound recorded by an input unit, a processor can apply a noise cancellation algorithm that considers as input the audio data produced by the auscultation microphone(s) and the audio data produced by the ambient microphone(s). The audio data may be digitized directly in the input unit, and then transmitted to the hub unit for synchronization. For example, by examining the audio data produced by the ambient microphone(s), the processor may discover which digital artifacts, if any, should be filtered from the audio data produced by the auscultation microphone(s).

Abstract: One aspect of the present application provides a computer-implemented method for generating a 3D digital model of gingiva, comprising: obtaining a 3D digital model of gingiva in a first state; and performing deformation process on the 3D digital model of the gingiva in the first state based on its deformation control points and deformation control points of a 3D digital model of crowns in a second state, to obtain a 3D digital model of the gingiva in the second state, wherein corresponding deformation control points of 3D digital models of the crowns and the gingiva in the same state coincide.

Abstract: Introduced here are electronic stethoscope systems designed to simultaneously monitor sounds originating from within a body under examination and the ambient environment. An electronic stethoscope system can include one or more input units that are connected to a hub unit. Each input unit may have at least one auscultation microphone and at least one ambient microphone. To improve the quality of sound recorded by an input unit, a processor can apply a noise cancellation algorithm that considers as input the audio data produced by the auscultation microphone(s) and the audio data produced by the ambient microphone(s). The audio data may be digitized directly in the input unit, and then transmitted to the hub unit for synchronization. For example, by examining the audio data produced by the ambient microphone(s), the processor may discover which digital artifacts, if any, should be filtered from the audio data produced by the auscultation microphone(s).

Abstract: Methods and systems for performing semantic scene completion of sparse 3D data are described. A frame of sparse 3D data is preprocessed into a sparse 3D tensor and a sparse 2D tensor. A partially completed 3D tensor is generated from the sparse 3D tensor using a 3D prediction network, and a semantically completed 2D tensor is generated from the sparse 2D tensor using a 2D prediction network. The partially completed 3D tensor is completed to obtain a semantically completed 3D tensor by assigning a given class label, which has been assigned to a given pixel in the semantically completed 2D tensor, to a voxel at a corresponding x-y coordinate in the partially completed 3D tensor.

Abstract: Disclosed are an electronic timer socket and a timing setting method thereof. A user-defined timing mode, a Daily timing mode, a 7-Day cycle timing mode, a CTD timing mode and a normally-on/normally-off timing mode are provided through key setting, and a two-way socket circuit may be independently controlled to be turned on or off through the above modes, so that the advantages of simple setting and convenience for operation are achieved, the learning difficulty of using products is reduced, and greater convenience is brought for the public to use. Through the above-mentioned modes and a multi-mode parallel or serial discrimination rule, the two-way socket circuit may be independently controlled to be turned on or off, which has the advantages of simple setting and convenience for operation, making the use of the timer socket more flexible and more suitable for more scenarios.

Abstract: System and method for semantic segmentation of point clouds. The method may include: generating, via a first neural network, a birds-eye-view (BEV) image of the environment from the aggregated point cloud; generating, via a second neural network, a labelled BEV image from the BEV image, wherein each pixel in the labelled BEV image is associated with a class label from a set of class labels; generating a BEV feature map; and generating, via a third neural network, the road surface segmentation map in the form of a refined labelled BEV image based on the labelled BEV image by smoothing the labelled BEV image using the BEV feature map, wherein each pixel in the refined labelled BEV image is associated with a class label from the set of class labels.

Abstract: A method of measuring cross-correlation or similarity between input features and filters of neural networks using an RRAM-crossbar architecture to carry out addition/subtraction-based neural networks for in-memory computing. The correlation calculations use L1 norm operations of AdderNet. The RCM structure of the RRAM-cross bar has storage and computing collocated, such that processing is done in the analog domain with low power, low latency and small area. In addition, the impact due to the nonidealities of RRAM device can be alleviated by the implicit ratio-based feature of the structure.

Abstract: Introduced here are computer programs and associated computer-implemented techniques for deriving insights into the health of patients through analysis of audio data generated by electronic stethoscope systems. A diagnostic platform may be responsible for examining the audio data generated by an electronic stethoscope system so as to gain insights into the health of a patient. The diagnostic platform may employ heuristics, algorithms, or models that rely on machine learning or artificial intelligence to perform auscultation in a manner that significantly outperforms traditional approaches that rely on visual analysis by a healthcare professional.

Abstract: Introduced here are computer programs and associated computer-implemented techniques for deriving insights into the health of patients through analysis of audio data generated by electronic stethoscope systems. A diagnostic platform may be responsible for examining the audio data generated by an electronic stethoscope system so as to gain insights into the health of a patient. The diagnostic platform may employ heuristics, algorithms, or models that rely on machine learning or artificial intelligence to perform auscultation in a manner that significantly outperforms traditional approaches that rely on visual analysis by a healthcare professional.

Abstract: A system and method for processing a 3D point cloud to generate a segmented point cloud in real time are disclosed, the method includes: receiving a sparse 3D point cloud captured by a detection and ranging sensor mounted to a vehicle, the 3D point cloud comprising a plurality of data points, each data point in the 3D point cloud having a set of coordinates in a coordinate system of the detection and ranging sensor; generating, from the 3D point cloud, a range map comprising a plurality of elements, each of the plurality of data points of the 3D point cloud occupying a respective element of the plurality of elements; labelling the data point in each respective element of the range map as one of a pole-like data point or a vertical-plane-like data point; and generating the segmented point cloud including one or more of the labeled data points.

Abstract: Introduced here are computer programs and associated computer-implemented techniques for deriving insights into the health of patients through analysis of audio data generated by electronic stethoscope systems. A diagnostic platform may be responsible for examining the audio data generated by an electronic stethoscope system so as to gain insights into the health of a patient. The diagnostic platform may employ heuristics, algorithms, or models that rely on machine learning or artificial intelligence to perform auscultation in a manner that significantly outperforms traditional approaches that rely on visual analysis by a healthcare professional.

Abstract: System and method for semantic segmentation of point clouds. The method may include: generating, via a first neural network, a birds-eye-view (BEV) image of the environment from the aggregated point cloud; generating, via a second neural network, a labelled BEV image from the BEV image, wherein each pixel in the labelled BEV image is associated with a class label from a set of class labels; generating a BEV feature map; and generating, via a third neural network, the road surface segmentation map in the form of a refined labelled BEV image based on the labelled BEV image by smoothing the labelled BEV image using the BEV feature map, wherein each pixel in the refined labelled BEV image is associated with a class label from the set of class labels.

Abstract: Methods and systems for allowing for techniques of automatically targeting focus and enlarging images upon detection that a receiver of shared video content is using a smaller screen. The presenter may indicate a focus point on the displayed video being shared to indicate where the targeted focusing and enlarging should be centered. A focus area and an enlargement factor may be calculated based on a difference in screen size between a presenter device and a destination device. The focus area and enlargement factor may be calculated such that elements of the shared video content are substantially the same size to a viewer of the destination device as they appear to the presenter on their presenter device. Different destination devices may have different focus areas and enlargement factors.

Abstract: Methods and systems for performing semantic scene completion of sparse 3D data are described. A frame of sparse 3D data is preprocessed into a sparse 3D tensor and a sparse 2D tensor. A partially completed 3D tensor is generated from the sparse 3D tensor using a 3D prediction network, and a semantically completed 2D tensor is generated from the sparse 2D tensor using a 2D prediction network. The partially completed 3D tensor is completed to obtain a semantically completed 3D tensor by assigning a given class label, which has been assigned to a given pixel in the semantically completed 2D tensor, to a voxel at a corresponding x-y coordinate in the partially completed 3D tensor.

Abstract: A computer-implemented method, a computing system, and a non-transitory machine-readable medium for semantic segmentation of a point cloud frame are provided. Point cloud frames including a target point cloud frame are received. For each sequence of a sliding set of sequences of point cloud frames, the sequence including the target point cloud frame, each point cloud frame in the sequence of point cloud frames is semantically segmented to apply semantic labels to points. A most prevalent semantic label is determined for each point in the target point cloud frame across the sliding set of sequences of point cloud frames.

Abstract: Methods and systems for computing a surface normal vector for a point cloud are disclosed. A sparse range map is generated by projecting the point cloud. A vertical kernel and a horizontal kernel are used to generate a vertical kernel range map and a horizontal kernel range map from the sparse range map. To complete an empty entry in the sparse range map, a guidance range map is used compute a vertical gradient and a horizontal gradient. The empty entry is filled using a corresponding entry in the vertical kernel range map or in the horizontal kernel range map, based on a comparison between the vertical gradient and the horizontal gradient. A surface normal vector is computed using the completed dense range map.