Abstract: Methods, systems, and devices for personalized (e.g., user specific) eye openness estimation are described. A network model (e.g., a convolutional neural network) may be trained using a set of synthetic eye openness image data (e.g., synthetic face images with known degrees or percentages of eye openness) and a set of real eye openness image data (e.g., facial images of real persons that are annotated as either open eyed or closed eyed). A device may estimate, using the network model, a multi-stage eye openness level (e.g., a percentage or degree to which an eye is open) of a user based on captured real time eye openness image data. The degree of eye openness estimated by the network model may then be compared to an eye size of the user (e.g., a user specific maximum eye size), and a user specific eye openness level may be estimated based on the comparison.

Abstract: Methods, systems, and devices for object recognition are described. Generally, the described techniques provide for a compact and efficient convolutional neural network (CNN) model for facial recognition. The proposed techniques relate to a light model with a set of layers of convolution and one fully connected layer for feature representation. A new building block of for each convolution layer is proposed. A maximum feature map (MFM) operation may be employed to reduce channels (e.g., by combining two or more channels via maximum feature selection within the channels). Depth-wise separable convolution may be employed for computation reduction (e.g., reduction of convolution computation). Batch normalization may be applied to normalize the output of the convolution layers and the fully connected layer (e.g., to prevent overfitting). The described techniques provide a compact and efficient CNN model which can be used for efficient and effective face recognition.

Abstract: Methods, systems, and devices for image processing are described. The method may include identifying a face in a first image based on identifying one or more biometric features of the face, determining an angular direction of one or more pixels of the identified face, identifying an anchor point on the identified face, sorting each of one or more pixels of the identified face into one of a set of pixel bins based on a combination of the determined angular direction of the pixel and a distance between the pixel and the identified anchor point, and outputting an indication of authenticity associated with the face based on a number of pixels in each bin.

Abstract: Methods, systems, and devices for object recognition are described. A device may generate a subspace based at least in part on a set of representative feature vectors for an object. The device may obtain an array of pixels representing an image. The device may determine a probe feature vector for the image by applying a convolutional operation to the array of pixels. The device may create a reconstructed feature vector in the subspace based at least in part on the set of representative feature vectors and the probe feature vector. The device may compare the reconstructed feature vector and the probe feature vector and recognize the object in the image based at least in part on the comparison. For example, the described techniques may support pose invariant facial recognition or other such object recognition applications.

Abstract: A method performed by an electronic device is described. The method includes incrementally adding a current node to a graph. The method also includes incrementally determining a respective adaptive edge threshold for each candidate edge between the current node and one or more candidate neighbor nodes. The method further includes determining whether to accept or reject each candidate edge based on each respective adaptive edge threshold. The method additionally includes performing refining based on the graph to produce refined data. The method also includes producing a three-dimensional (3D) model based on the refined data.

Abstract: A method is described. The method includes determining normalized radiance of an image sequence based on a camera response function (CRF). The method also includes determining one or more reliability images of the image sequence based on a reliability function corresponding to the CRF. The method further includes extracting features based on the normalized radiance of the image sequence. The method additionally includes optimizing a model based on the extracted features and the reliability images.

Abstract: A method includes receiving a user input (e.g., a one-touch user input), performing segmentation to generate multiple candidate regions of interest (ROIs) in response to the user input, and performing ROI fusion to generate a final ROI (e.g., for a computer vision application). In some cases, the segmentation may include motion-based segmentation, color-based segmentation, or a combination thereof. Further, in some cases, the ROI fusion may include intraframe (or spatial) ROI fusion, temporal ROI fusion, or a combination thereof.

Abstract: Techniques and systems are provided for authenticating a user of a device. For example, input biometric data associated with a person can be obtained. A similarity score for the input biometric data can be determined by comparing the input biometric data to a set of templates that include reference biometric data associated with the user. The similarity score can be compared to an authentication threshold. The person is authenticated as the user when the similarity score is greater than the authentication threshold. The similarity score can also be compared to a learning threshold that is greater than the authentication threshold. A new template including features of the input biometric data is saved for the user when the similarity score is less than the learning threshold and greater than the authentication threshold.

Abstract: A method performed by an electronic device is described. The method includes receiving a set of frames. The set of frames describes a moving three-dimensional (3D) object. The method also includes registering the set of frames based on a canonical model. The canonical model includes geometric information and optical information. The method additionally includes fusing frame information of each frame to the canonical model based on the registration. The method further includes reconstructing the 3D object based on the canonical model.

Abstract: A depth based scanning system can be configured to determine whether pixel depth values of a depth map are within a depth range; determine a component of the depth map comprised of connected pixels each with a depth value within the depth range; replace the depth values of any pixels of the depth map that are not connected pixels; determine whether each pixel of the connected pixels of the component has at least a threshold number of neighboring pixels that have a depth value within the depth range; and for each pixel of the connected pixels of the component, if the pixel is determined to have at least the threshold number of neighboring pixels, replace its depth value with a filtered depth value that is based on the depth values of the neighboring pixels that have a depth value within the depth range.

Abstract: Techniques and systems are provided for performing object verification using radar images. For example, a first radar image and a second radar image are obtained, and features are extracted from the first radar image and the second radar image. A similarity is determined between an object represented by the first radar image and an object represented by the second radar image based on the features extracted from the first radar image and the features extracted from the second radar image. A determined similarity between these two sets of features is used to determine whether the object represented by the first radar image matches the object represented by the second radar image. Distances between the features in the two radar images can optionally also be compared and used to determine object similarity. The objects in the radar images may optionally be faces.

Abstract: Embodiments described herein can address these and other issues by using radar machine learning to address the radio frequency (RF) to perform object identification, including facial recognition. In particular, embodiments may obtain IQ samples by transmitting and receiving a plurality of data packets with a respective plurality of transmitter antenna elements and receiver antenna elements. I/Q samples indicative of a channel impulse responses of an identification region obtained from the transmission and reception of the plurality of data packets may then be used to identify, with an autoencoder, a physical object in the identification region.

Abstract: Embodiments described herein can address these and other issues by using radar machine learning to address the radio frequency (RF) to perform object identification, including facial recognition. In particular, embodiments may obtain IQ samples by transmitting and receiving a plurality of data packets with a respective plurality of transmitter antenna elements and receiver antenna elements, where each data packet of the plurality of data packets comprises one or more complementary pairs of Golay sequences. I/Q samples indicative of a channel impulse responses of an identification region obtained from the transmission and reception of the plurality of data packets may then be used to identify, with a random forest model, a physical object in the identification region.

Abstract: An apparatus includes a first sensor configured to generate first sensor data. The first sensor data is related to an occupant of a vehicle. The apparatus further includes a depth sensor and a processor. The depth sensor is configured to generate data corresponding to a volume associated with at least a portion of the occupant. The processor is configured to receive the first sensor data and to activate the depth sensor based on the first sensor data.

Abstract: Methods, systems, and devices for object recognition are described. A device may generate a subspace based at least in part on a set of representative feature vectors for an object. The device may obtain an array of pixels representing an image. The device may determine a probe feature vector for the image by applying a convolutional operation to the array of pixels. The device may create a reconstructed feature vector in the subspace based at least in part on the set of representative feature vectors and the probe feature vector. The device may compare the reconstructed feature vector and the probe feature vector and recognize the object in the image based at least in part on the comparison. For example, the described techniques may support pose invariant facial recognition or other such object recognition applications.

Abstract: In various implementations, object tracking in a video content analysis system can be augmented with an image-based object re-identification system (e.g., for person re-identification or re-identification of other objects) to improve object tracking results for objects moving in a scene. The object re-identification system can use image recognition principles, which can be enhanced by considering data provided by object trackers that can be output by an object traffic system. In a testing stage, the object re-identification system can selectively test object trackers against object models. For most input video frames, not all object trackers need be tested against all object models. Additionally, different types of object trackers can be tested differently, so that a context provided by each object tracker can be considered. In a training stage, object models can also be selectively updated.

Abstract: Methods, systems, and devices for object recognition are described. Generally, the described techniques provide for a compact and efficient convolutional neural network (CNN) model for facial recognition. The proposed techniques relate to a light model with a set of layers of convolution and one fully connected layer for feature representation. A new building block of for each convolution layer is proposed. A maximum feature map (MFM) operation may be employed to reduce channels (e.g., by combining two or more channels via maximum feature selection within the channels). Depth-wise separable convolution may be employed for computation reduction (e.g., reduction of convolution computation). Batch normalization may be applied to normalize the output of the convolution layers and the fully connected layer (e.g., to prevent overfitting). The described techniques provide a compact and efficient CNN model which can be used for efficient and effective face recognition.

Abstract: A depth based scanning system can be configured to determine whether pixel depth values of a depth map are within a depth range; determine a component of the depth map comprised of connected pixels each with a depth value within the depth range; replace the depth values of any pixels of the depth map that are not connected pixels; determine whether each pixel of the connected pixels of the component has at least a threshold number of neighboring pixels that have a depth value within the depth range; and for each pixel of the connected pixels of the component, if the pixel is determined to have at least the threshold number of neighboring pixels, replace its depth value with a filtered depth value that is based on the depth values of the neighboring pixels that have a depth value within the depth range.

Abstract: A method performed by an electronic device is described. The method includes incrementally adding a current node to a graph. The method also includes incrementally determining a respective adaptive edge threshold for each candidate edge between the current node and one or more candidate neighbor nodes. The method further includes determining whether to accept or reject each candidate edge based on each respective adaptive edge threshold. The method additionally includes performing refining based on the graph to produce refined data. The method also includes producing a three-dimensional (3D) model based on the refined data.

Abstract: A method is described. The method includes determining normalized radiance of an image sequence based on a camera response function (CRF). The method also includes determining one or more reliability images of the image sequence based on a reliability function corresponding to the CRF. The method further includes extracting features based on the normalized radiance of the image sequence. The method additionally includes optimizing a model based on the extracted features and the reliability images.