Abstract: Technology described herein includes a method that includes obtaining a set of color-coded sequences, each of which includes a sequence of colors. Each color-coded sequence has auto-correlation properties characterized by a merit factor larger than a first predetermined threshold, and cross-correlation properties among the color-coded sequences characterized by a demerit factor lower than a second predetermined threshold. A color-coded sequence is randomly selected from the set of color-coded sequences. A subject is illuminated in accordance with the sequence of colors in the selected color-coded sequence. A sequence of images of the subject are captured, and are temporally synchronized with illumination by the color-coded sequence. A filtered response image is generated from the sequence of images by a matched filtering process. Based on the filtered response image, it is determined that the subject is an alternative representation of a live person. In response, access to a secure system is prevented.

Abstract: An adaptive spoofing method includes obtaining multiple images of a spoof representation of a human user. The multiple images are captured automatically by an image capture device, and each of the multiple images is captured at a different relative position between the image capture device and the spoof representation. The method also includes: executing a biometric authentication process separately on each of at least a subset of the multiple images; determining that the biometric authentication process authenticates the human user based on at least a first image from the subset; and identifying a relative position between the image capture device and the spoof representation corresponding to the first image as a failure condition associated with the biometric authentication process.

Abstract: An adaptive spoofing method includes: causing a first display device to display a spoof representation of a human user; and obtaining multiple images of the spoof representation. The multiple images are captured automatically by an image capture device, each of the multiple images captures a different corresponding configuration of the spoof representation displayed on the first display device. The method also includes: executing a biometric authentication process separately on each of at least a subset of the multiple images; determining that the biometric authentication process authenticates the human user based on at least a first image from the subset; and identifying a configuration of the spoof representation corresponding to the first image as a failure condition associated with the biometric authentication process.

Abstract: Technology described herein includes a method that includes receiving, at one or more processing devices at one or more locations, one or more image frames; receiving, at the one or more processing devices, one or more signals representing outputs of one or more light sensors of a device; estimating, by the one or more processing devices based on the one or more image frames, one or more illuminance values; determining, by the one or more processing devices, that a degree of correlation between (i) a first illuminance represented by the one or more illuminance values and (ii) a second illuminance represented by the one or more signals fails to satisfy a threshold condition; and in response to determining that the degree of correlation fails to satisfy the threshold condition, determining, by the one or more processing devices, presence of an adverse condition associated with the device.

Abstract: An image transformation method includes: causing a first display device to display one or more calibration patterns. One or more first images include a corresponding calibration pattern of the one or more calibration patterns. The one or more first images are captured by an image capture device disposed at a non-zero capture angle with respect to the first display device. The method also includes determining based on the one or more first images, a transformation to be applied to account for the non-zero capture angle; obtaining a second image of a human user; applying the transformation to the second image to generate a transformed spoof representation of the human user for capture by the image capture device being held at the non-zero capture angle with respect to the first display device; and causing the first display device to display the transformed spoof representation.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium for red eye detection are provided. In one aspect, a system includes an image acquisition device, first and second illuminators, and at least one processor. The first illuminator is arranged closer to the image acquisition device than the second illuminator. The image acquisition device is configured to capture a first facial image of a face of a subject with the first illuminator being on and the second illuminator being off and a second facial image of the face of the subject with the second illuminator being on and the first illuminator being off. The processor can process the first facial image based on the second facial image to determine whether at least one eye of the subject is live by determining that the first facial image includes a red eye reflection from the at least one eye.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer-readable storage media, are provided. In one aspect, a first image of an eye region of a subject is received, the first image captured under illumination by electromagnetic radiation having a wavelength in the 850 nm-1100 nm range. Based on the first image, a metric indicative of differential absorption of the electromagnetic radiation by an iris of the subject and a sclera of the subject is determined. It is identified, based on the metric, that the subject is a spoof representation of a human. Responsive to identifying that the subject is a spoof representation of a human, access to a secure system is prevented.

Abstract: Technology described herein includes a method that includes obtaining a set of color-coded sequences, each of which includes a sequence of colors. Each color-coded sequence has auto-correlation properties characterized by a merit factor larger than a first predetermined threshold, and cross-correlation properties among the color-coded sequences characterized by a demerit factor lower than a second predetermined threshold. A color-coded sequence is randomly selected from the set of color-coded sequences. A subject is illuminated in accordance with the sequence of colors in the selected color-coded sequence. A sequence of images of the subject are captured, and are temporally synchronized with illumination by the color-coded sequence. A filtered response image is generated from the sequence of images by a matched filtering process. Based on the filtered response image, it is determined that the subject is an alternative representation of a live person. In response, access to a secure system is prevented.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium for red eye detection are provided. In one aspect, a system includes an image acquisition device, first and second illuminators, and at least one processor. The first illuminator is arranged closer to the image acquisition device than the second illuminator. The image acquisition device is configured to capture a first facial image of a face of a subject with the first illuminator being on and the second illuminator being off and a second facial image of the face of the subject with the second illuminator being on and the first illuminator being off. The processor can process the first facial image based on the second facial image to determine whether at least one eye of the subject is live by determining that the first facial image includes a red eye reflection from the at least one eye.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for biometric identity authentication. In one aspect, a method comprises: determining: (i) a plurality of first interest points in a first image of an eye of a first user, and (ii) a plurality of second interest points in a second image of an eye of a second user; determining a plurality of interest point pairs; processing a network input comprising the interest point pairs using an inlier neural network to generate a network output comprising a plurality of inlier scores; and determining whether an identity of the first user matches an identity of the second user based on the inlier scores.

Abstract: Computer-implemented methods and systems for image security in computer vision applications are described. A shared secret is stored at an image sensor module and a trusted element. The shared secret includes a device-specific secret value and a definition of a plurality of pixel locations. An image is received at the image sensor module, which generates a watermark based on the secret value and applies the watermark to the image, based on the pixel locations, to create a watermarked image. The watermarked image is received at the trusted element, which validates the watermarked image based on the watermark.

Abstract: The technology described in this document can be embodied in a method for identifying a user of secured system. The method includes transmitting, using a first transceiver in an array of multiple transceivers, a first electromagnetic signal in the microwave range. An electromagnetic signal representing a first response of a portion of a subject to the first electromagnetic signal is received, by at least one transceiver of the array of multiple transceivers. The subject is identified based at least in part on one or more parameters of the first response.

Abstract: In a feature extraction and pattern matching system, image sharpening can enable vascular point detection (VPD) for detecting points of interest from visible vasculature of the eye. Pattern Histograms of Extended Multi-Radii Local Binary Patterns and/or Pattern Histograms of Extended Multi-Radii Center Symmetric Local Binary Patterns can provide description of portions of images surrounding a point of interest, and enrollment and verification templates can be generated using points detected via VPD and the corresponding descriptors. Inlier point pairs can be selected from the enrollment and verification templates, and a first match score indicating similarity of the two templates can be computed based on the number of inlier point pairs and one or more parameters of a transform selected by the inlier detection. A second match score can be computed by applying the selected transform, and either or both scores can be used to authenticate the user.

Abstract: In a feature extraction and pattern matching system, image sharpening can enable vascular point detection (VPD) for detecting points of interest from visible vasculature of the eye. Pattern Histograms of Extended Multi-Radii Local Binary Patterns and/or Pattern Histograms of Extended Multi-Radii Center Symmetric Local Binary Patterns can provide description of portions of images surrounding a point of interest, and enrollment and verification templates can be generated using points detected via VPD and the corresponding descriptors. Inlier point pairs can be selected from the enrollment and verification templates, and a first match score indicating similarity of the two templates can be computed based on the number of inlier point pairs and one or more parameters of a transform selected by the inlier detection. A second match score can be computed by applying the selected transform, and either or both scores can be used to authenticate the user.

Abstract: In a feature extraction and pattern matching system, image sharpening can enable vascular point detection (VPD) for detecting points of interest from visible vasculature of the eye. Pattern Histograms of Extended Multi-Radii Local Binary Patterns and/or Pattern Histograms of Extended Multi-Radii Center Symmetric Local Binary Patterns can provide description of portions of images surrounding a point of interest, and enrollment and verification templates can be generated using points detected via VPD and the corresponding descriptors. Inlier point pairs can be selected from the enrollment and verification templates, and a first match score indicating similarity of the two templates can be computed based on the number of inlier point pairs and one or more parameters of a transform selected by the Inlier detection. A second match score can be computed by applying the selected transform, and either or both scores can be used to authenticate the user.

Abstract: Computer-implemented methods and systems for liveness analysis using proximity sensors are described. Reflective strength readings from a proximity sensor, such as an infrared proximity sensor incorporated in a mobile device, are received. A liveness measure is determined based on the one or more readings and a reflection threshold, and, based on the liveness measure, a determination is made whether a target in range of the proximity sensor is likely to be a live human.

Abstract: Systems and methods for electronically leashing a user to a mobile device. A user is authenticated on the mobile device and initial sensor data (e.g., radio signal readings, accelerometer readings, image/video, audio) is collected. Based on a timer or other triggering event, additional sensor data is captured and evaluated. Based on the evaluation of the sensor data, a value representing a likelihood of whether the device remains in possession of the user is determined. Upon determining that this value is less than a threshold, the user is required to reauthenticate on the mobile device to further engage with the device or particular features on the device.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for biometric identity authentication. In one aspect, a method comprises: determining: (i) a plurality of first interest points in a first image of an eye of a first user, and (ii) a plurality of second interest points in a second image of an eye of a second user; determining a plurality of interest point pairs; processing a network input comprising the interest point pairs using an inlier neural network to generate a network output comprising a plurality of inlier scores; and determining whether an identity of the first user matches an identity of the second user based on the inlier scores.

Abstract: A biometric template created at a user device is divided into portions that are distributed among members of a trusted circle and, optionally, a remote storage service. When the user associated with the biometric template attempts to reauthenticate on a different user device, live identity information is captured and transmitted to trusted circle members. The members confirm the identity of the user and provide the biometric template portions to the different device for reconstruction of the original template. The user can then biometrically reauthenticate using the reconstructed template.

Abstract: Systems and methods for electronically leashing a user to a mobile device. A user is authenticated on the mobile device and initial sensor data (e.g., radio signal readings, accelerometer readings, image/video, audio) is collected. Based on a timer or other triggering event, additional sensor data is captured and evaluated. Based on the evaluation of the sensor data, a value representing a likelihood of whether the device remains in possession of the user is determined. Upon determining that this value is less than a threshold, the user is required to reauthenticate on the mobile device to further engage with the device or particular features on the device.