Abstract: This document describes systems and techniques that enable continuous personalization of face authentication. In aspects, an authentication system associated with a network includes an authentication manager. The authentication manager receives an embedding representing image data associated with a user's face. The authentication manager generates a confidence score based on the embedding. Further, the authentication manager updates previously enrolled embeddings with the embedding based on the confidence score, the embedding meeting a clustering confidence threshold. Through such a technique, the authentication manager can alter the previously enrolled embeddings by which a future embedding is used to authenticate the user's face. By so doing, the techniques may provide more-accurate and successful user authentication over time.

Abstract: Systems and methods for training a generative neural radiance field model can include geometric regularization. Geometric regularization can involve the utilization of reference geometry data and/or an output of a surface prediction model. The geometry regularization can train the generative neural radiance field model to mitigate artifact generation by limiting a distribution considered for color value prediction and density value prediction to a range associated with a realistic geometry range.

Abstract: A method includes obtaining dual-pixel image data that represents an object and includes a first sub-image and a second sub-image, and generating (i) a first feature map based on the first sub-image and (ii) a second feature map based on the second sub-image. The method also includes generating a correlation volume by determining, for each respective offset of a plurality of offsets between the first feature map and the second feature map, pixel-wise similarities between (i) the first feature map and (ii) the second feature map offset from the first feature map by the respective offset. The method further includes determining, by an anti-spoofing model and based on the correlation volume, a spoofing value indicative of a likelihood that the object represented by the dual-pixel image data is being spoofed.

Abstract: A method includes receiving data indicative of an image of a face of an unknown user of the computing device while the computing device is in a reduced access mode locked state. The method also includes determining whether the unknown user is the known user by at least comparing the image of the face of the unknown user to one or more images of a plurality of images of a face of a known user of the computing device. The method further includes setting the computing device to an increased access mode in response to determining that the unknown user is the known user.

Abstract: Provided is a multi-scale model ensemble for detection of objects in images. The model ensemble can be applied, for example, in the context of performing object identification activities, such as positively identifying desired objects in image data or video data using a variety of different crop levels.

Abstract: Provided are systems and methods for detecting an object in an image. The method can include receiving an input image and analyzing the input image using an image segmentation model to identify one or more indicative areas within the input image, the one or more indicative areas being indicative of one or more objects within the input image. The method can also include analyzing the one or more indicative areas of the input image using a convolutional model to generate at least one label for at least one portion of the one or more indicative areas of the input image, the label indicating whether a specific object is identified within the input image, and performing at least one action based on the at least one label for the at least one portion.

Abstract: Systems and methods of the present disclosure are directed to a computer-implemented method. The method can include obtaining a first image depicting a first object and a second image depicting a second object, wherein the first object comprises a first feature set and the second object comprises a second feature set. The method can include processing the first image with a machine-learned image transformation model comprising a plurality of model channels to obtain a first channel mapping indicative of a mapping between the plurality of model channels and the first feature set. The method can include processing the second image with the model to obtain a second channel mapping indicative of a mapping between the plurality of model channels and the second feature set. The method can include generating an interpolation vector for a selected feature.

Abstract: The present disclosure provides improved methods for learning a generative model with limited training data, by leveraging a pre-trained GAN model from a related domain and adapting it to the new domain given a set of target examples from the new or target domain.

Abstract: A method may include obtaining an infrared image of an object and determining a difference of Gaussian image that represents features of the infrared image that have spatial frequencies within a spatial frequency range defined by a first Gaussian operator and a second Gaussian operator. The method may also include identifying one or more blob regions within the difference of Gaussian image. Each blob region of the one or more blob regions includes a region of connected pixels in the difference of Gaussian image. The method may further include, based on identifying the one or more blob regions within the difference of Gaussian image, determining that the infrared image represents the object illuminated by a pattern projected onto the object by an infrared projector.

Abstract: Systems and methods of the present disclosure are directed to a computer-implemented method. The method can include obtaining a first image depicting a first object and a second image depicting a second object, wherein the first object comprises a first feature set and the second object comprises a second feature set. The method can include processing the first image with a machine-learned image transformation model comprising a plurality of model channels to obtain a first channel mapping indicative of a mapping between the plurality of model channels and the first feature set. The method can include processing the second image with the model to obtain a second channel mapping indicative of a mapping between the plurality of model channels and the second feature set. The method can include generating an interpolation vector for a selected feature.

Abstract: A method includes receiving data indicative of an image of a face of an unknown user of the computing device while the computing device is in a reduced access mode locked state. The method also includes determining whether the unknown user is the known user by at least comparing the image of the face of the unknown user to one or more images of a plurality of images of a face of a known user of the computing device. The method further includes setting the computing device to an increased access mode in response to determining that the unknown user is the known user.

Abstract: A method may include obtaining an infrared image of an object and determining a difference of Gaussian image that represents features of the infrared image that have spatial frequencies within a spatial frequency range defined by a first Gaussian operator and a second Gaussian operator. The method may also include identifying one or more blob regions within the difference of Gaussian image. Each blob region of the one or more blob regions includes a region of connected pixels in the difference of Gaussian image. The method may further include, based on identifying the one or more blob regions within the difference of Gaussian image, determining that the infrared image represents the object illuminated by a pattern projected onto the object by an infrared projector.

Abstract: Methods are provided to determine a quality score for depth map. The quality score is calculated from metrics that detect artifacts or other inaccuracies in the depth map such as flat patches, artifactual edges, and patchy regions. A flatness metric detects regions of neighboring pixels that have substantially the same depth value. A jaggedness metric detects hard edges or other discontinuities. A patchiness metric detects regions that are wholly enclosed by an edge and that have sub-threshold areas. The individual metrics are normalized and combined to determine an overall quality score for the depth map. The quality score can then be compared to one or more thresholds to determine a quality label for the depth map. Such a quality label can then be used to unlock a device, to invalidate an unlock attempt, to recalibrate a depth sensor, or to perform some other operations.

Abstract: Methods are provided to determine a quality score for depth map. The quality score is calculated from metrics that detect artifacts or other inaccuracies in the depth map such as flat patches, artifactual edges, and patchy regions. A flatness metric detects regions of neighboring pixels that have substantially the same depth value. A jaggedness metric detects hard edges or other discontinuities. A patchiness metric detects regions that are wholly enclosed by an edge and that have sub-threshold areas. The individual metrics are normalized and combined to determine an overall quality score for the depth map. The quality score can then be compared to one or more thresholds to determine a quality label for the depth map. Such a quality label can then be used to unlock a device, to invalidate an unlock attempt, to recalibrate a depth sensor, or to perform some other operations.

Abstract: An example method includes establishing, by a mobile computing device, a connection with a vehicle computing system of a vehicle, receiving, from the vehicle computing system, feature data associated with at least one image of a face of a user of the vehicle, wherein the at least one image of the face is captured by an image capture device included in the vehicle, determining, based on a comparison between the feature data associated with the at least one image of the face of the user and feature data of at least one image of a face of a previously enrolled user, a match between the user of the vehicle and the previously enrolled user, authenticating, based on the match, the user of the vehicle, and sending, to the vehicle computing system, authentication data for the user of the vehicle, wherein the authentication data is indicative of the match.