QUALCOMM TECHNOLOGIES, INC.

Certain aspects of the present disclosure provide a method for performing machine learning, comprising: determining a plurality of vertices in a neighborhood associated with a mesh including a target vertex; determining a linear transformation configured to parallel transport signals along all edges in the mesh to the target vertex; applying the linear transformation to the plurality of vertices in the neighborhood to form a combined signal at the target vertex; determining a set of basis filters; linearly combining the basis filters using a set of learned parameters to form a gauge equivariant convolution filter, wherein the gauge equivariant convolution filter is constrained to maintain gauge equivariance; applying the gauge equivariant convolution filter to the combined signal to form an intermediate output; and applying a nonlinearity to the intermediate output to form a convolution output.

A computer-implemented method for contrastive object representation from temporal data using an artificial neural network (ANN) includes receiving, by the ANN, a video. The video comprises a temporal sequence of frames including images of one or more objects. The ANN generates object representations corresponding to the one or more objects based on temporal data of multiple frames of the temporal sequence of frames. The object representations are communicated to a receiver.

A package that includes a substrate, an integrated device coupled to the substrate, an encapsulation layer located over the substrate, at least one encapsulation layer interconnect located in the encapsulation layer, and a metal layer located over the encapsulation layer. The substrate includes at least one dielectric layer and a plurality of interconnects. The encapsulation layer interconnect is coupled to the substrate. The metal layer is configured as an electromagnetic interference (EMI) shield for the package. The metal layer is located over a backside of the integrated device.

Embodiments include methods, and processing devices for implementing the methods. Various embodiments may include calculating a batch softmax normalization factor using a plurality of logit values from a plurality of logits of a layer of a neural network, normalizing the plurality of logit values using the batch softmax normalization factor, and mapping each of the normalized plurality of logit values to one of a plurality of manifolds in a coordinate space. In some embodiments, each of the plurality of manifolds represents a number of labels to which a logit can be classified. In some embodiments, at least one of the plurality of manifolds represents a number of labels other than one label.

A microphone including a casing having a front wall, a back wall, and a side wall joining the front wall to the back wall, a transducer mounted to the front wall, the transducer including a substrate and a transducing element, the transducing element having a transducer acoustic compliance dependent on the transducing element dimensions, a back cavity cooperatively defined between the back wall, the side wall, and the transducer, the back cavity having a back cavity acoustic compliance. The transducing element is dimensioned such that the transducing element length matches a predetermined resonant frequency and the transducing element width, thickness, and elasticity produces a transducer acoustic compliance within a given range of the back cavity acoustic compliance.

Certain aspects of the present disclosure provide techniques for pose estimation for three-dimensional object reconstruction. In one example, a method, includes receiving image data, wherein the image data comprises a plurality of images taken from varying poses; identifying one or more pairs of spatially related images within the plurality of images; generating a synchronization graph indicative of at least one similarity metric between the plurality of images, based at least in part on the identified one of more pairs of spatially related images; and estimating a pose of an object depicted in the plurality of images based on the synchronization graph.

A method for classifying a human-object interaction includes identifying a human-object interaction in the input. Context features of the input are identified. Each identified context feature is compared with the identified human-object interaction. An importance of the identified context feature is determined for the identified human-object interaction. The context feature is fused with the identified human-object interaction when the importance is greater than a threshold.

An acoustic device is described and includes an acoustic sensor element configured to sense acoustic energy and produce an output signal and a threshold detector circuit including a switch having an input coupled to the output of the acoustic sensor element to receive the output signal, a control port that receives a control signal, and first and second output ports, a first channel including an analog-to-digital converter that operates at a first power level a second analog-to-digital converter that operates at a second higher power level, relative to the first power level and a threshold level detector that receives an output from the first analog-to-digital converter to produce the control signal having a first state that causes the switch feed the output signal from the acoustic sensor element to the second analog-to-digital converter when the first digitized output signal meets a threshold criteria.

A method for generating a causal graph includes receiving a data set including observation data and intervention data corresponding to multiple variables. A probability distribution is determined for each variable based on the observation data. A likelihood of including each edge in the graph is computed based on the probability distribution and the intervention data. Each edge is a causal connection between variables of the multiple variables. The graph is generated based on the likelihood of including each edge. The graph may be updated by iteratively repeating the determination of the probability distribution and the computing of the likelihood of including each edge.

Image Signal Processing (ISP) optimization framework for computer vision applications is disclosed. The tuning of the ISP is performed automatically and presented as a nonlinear multi-objective optimization problem, followed by solving the problem using an evolutionary stochastic solver. An improved ISP of the embodiments of the invention includes at least features of search space reduction for reducing a number of ISP configurations, remapping the generated population to the reduced search space via mirroring, and global optimization function processing, which allow tuning all the blocks of the ISP at the same time instead of the prior art tuning of each ISP block separately. Also shown that an ISP tuned for image quality performs inferior compared with an ISP trained for a specific downstream image recognition task.