Abstract: Disclosed herein are various techniques for more precisely and reliably (a) positioning top and bottom border edges relative to textual content, (b) positioning left and right border edges relative to textual content, (c) positioning mixed edge borders relative to textual content, (d) positioning boundaries of a region of background shading that fall within borders of textual content, (e) positioning borders relative to textual content that spans columns, (f) positioning respective borders relative to discrete portions of textual content, (g) positioning collective borders relative to discrete, abutting portions of textual content, (h) applying stylized corner boundaries to a region of background shading, and (i) applying stylized corners to borders.

Abstract: There are several types of plenoptic devices having their proprietary file format. At present there is no standard supporting the acquisition and transmission of multi-dimensional information for an exhaustive over-view of the different parameters upon which a light-field depends. As such acquired light-field data for different cameras have a diversity of formats. The notion of pixel beam, which represents a volume occupied by a set of rays of light in an object space of an optical system of a camera is thus introduced. The method according to the invention enables to provide data representative of a collection of pixel beams describing a first optical system that is agnostic since these data are obtained by imaging the collection of pixel beams through a second optical system. Such data representative of a collection of pixel beams enable the generation of parameterized output images from which post-processing.

Abstract: Methods, apparatus and systems for wireless sensing based on deep learning are described.

Abstract: Systems and methods for eye tracking latency enhancements. An example head-mounted system obtains a first image of an eye of a user. The first image is provided as input to a machine learning model which has been trained to generate iris and pupil segmentation data given an image of an eye. A second image of the eye is obtained. A set of locations in the second image at which one or more glints are shown is detected based on iris segmentation data generated for the first image. A region of the second image at which the pupil of the eye of the user is shown is identified based on pupil segmentation data generated for the first image. A pose of the eye of the user is determined based on the detected set of glint locations in the second image and the identified region of the second image.

Abstract: Methods, apparatus and systems for radio-assisted signal estimation are described. In one example, a described system comprises: a sensor configured to obtain a baseband mixture signal in a venue; a transmitter configured to transmit a first radio signal through a wireless channel of the venue; a receiver configured to receive a second radio signal through the wireless channel; and a processor. The baseband mixture signal comprises a mixture of a first source signal and an additional signal. The first source signal is generated by a first motion of a first object in the venue. The second radio signal differs from the first radio signal due to the wireless channel and at least the first motion of the first object in the venue.

Abstract: A data processing method and apparatus, a device and a storage medium, where the method includes: obtaining first target result data according to an original ultrasonic echo signal, where the first target result data includes a related parameter of a detected object; performing feature extraction on the first target result data using a pre-trained feature extraction model to obtain second target result data; and performing corresponding processing on the detected object based on the second target result data. By performing feature extraction on the related parameter of the detected object using the pre-trained feature extraction model to obtain the second target result data, and further performing corresponding processing on the detected object based on the second target result data, thereby improving accuracy of judging the state of the detected object effectively.

Abstract: Systems and methods for a sample database are provided, where the sample database is for matching a target coating. In one embodiment, the system comprises a sample database stored on a storage device. The sample database includes a sample coating formula and a sample image feature with at least one sample coating formula linked to at least one sample image feature. At least one sample image feature includes a spatial micro-color analysis that includes a value determined by a sample pixel feature difference between at least two sample pixels.

Abstract: Example methods for adaptive radiotherapy treatment planning using deep learning engines are provided. One example method may comprise obtaining treatment image data associated with a first imaging modality and planning image data associated with a second imaging modality. The treatment image data may be acquired during a treatment phase of a patient. Also, planning image data associated with a second imaging modality may be acquired prior to the treatment phase to generate a treatment plan for the patient. The method may also comprise: in response to determination that an update of the treatment plan is required, processing, using the deep learning engine, the treatment image data and the planning image data to generate output data for updating the treatment plan.

Abstract: The present disclosure provides a method for correcting face distortion, an apparatus for correcting face distortion, an electronic device, and a storage medium. The method includes: performing face detection on an obtained image to determine a position of each face box included in the image; determining whether each face box is within a predetermined field of view range based on the position of the face box; and performing distortion correction on a face in a first face box in response to at least a part of the first face box being not within the predetermined field of view range, to generate a corrected image.

Abstract: Methods, apparatus and systems for enhanced wireless monitoring of vital signs are described. In one example, a described system comprises: a transmitter configured to transmit a wireless signal through a wireless channel of a venue; a receiver configured to receive the wireless signal through the wireless channel; and a processor. The received wireless signal differs from the transmitted wireless signal due to the wireless channel that is impacted by a periodic motion of a vital sign of an object in the venue. The processor is configured for: obtaining a time series of channel information (CI) of the wireless channel based on the received wireless signal, computing a two dimensional (2D) decomposition of the time series of CI (TSCI), enhancing the 2D decomposition, and monitoring the periodic motion of the vital sign based on the enhanced 2D decomposition.

Abstract: An image segmentation method includes obtaining target domain images and source domain images, and segmenting the source domain images and the target domain images by using a generative network in a first generative adversarial network. The method further includes segmenting the source domain images and the target domain images by using a generative network in a second generative adversarial network, and determining a first source domain image and a second source domain image according to source domain segmentation losses, and determining a first target domain image and a second target domain image according to target domain segmentation losses. The method also includes performing cross training on the first generative adversarial network and the second generative adversarial network to obtain a trained first generative adversarial network; and segmenting a to-be-segmented image based on the generative network in the trained first generative adversarial network.

Abstract: Methods, apparatus and systems for wireless monitoring and tracking are described. In one example, a described method of a wireless monitoring system comprises: transmitting a wireless signal that is impacted by a wireless multipath channel in a venue and a modulation of an object undergoing a motion in the venue, to obtain a set of channel information (CI) of the wireless multipath channel; performing a monitoring task by monitoring the object and the motion; determining a plurality of admissible system states of the wireless monitoring system, wherein each admissible system state is associated with a respective setting of at least one of: the wireless signal, a series of sounding signals in the wireless signal, or the monitoring task; choosing one of the admissible system states to be a system state of the wireless monitoring system based on the monitoring task; and applying a setting associated with the chosen admissible system state to the wireless monitoring system.

Abstract: Provided is an accelerator provided in an electronic device and configured to perform simultaneous localization and mapping (SLAM), the accelerator including a factor graph database, a memory, and a back-end processor, wherein the back-end processor is configured to receive a first piece of data corresponding to map points and camera positions from the factor graph database, convert the received first piece of data into a matrix for the map points and a matrix for the camera positions, store, in the memory, results obtained by performing an optimization calculation on the matrix for the map points and a matrix for at least one camera position, among the camera positions, corresponding to the map points, and obtain a second piece of data optimized with respect to the first piece of data based on the results stored in the memory.

Abstract: Methods, apparatus and systems for radio-based voice activity detection are described. In one example, a described system comprises: a transmitter configured to transmit a radio signal through a wireless channel of a venue; a receiver configured to receive the radio signal through the wireless channel, wherein the wireless channel is impacted by a voice activity of a target voice source in the venue; and a processor. The processor is configured for: computing a time series of channel information (CI) of the wireless channel based on the radio signal, and detecting the voice activity of the target voice source based on the time series of CI (TSCI) of the wireless channel, without using any media signal.

Abstract: A method that includes extracting image features of a document image, executing an optical character recognition (OCR) engine on the document image to obtain OCR output, and extracting OCR features from the OCR output. The method further includes executing an anomaly detection model using features including the OCR features and the image features to generate anomaly score, and presenting anomaly score.

Abstract: A neural network pruning system can sparsely prune neural network models using an optimizer based approach that is agnostic to the model architecture being pruned. The neural network pruning system can prune by operating on the parameter vector of the full model and the gradient vector of the loss function with respect to the model parameters. The neural network pruning system can iteratively update parameters based on the gradients, while zeroing out as many parameters as possible based a preconfigured penalty.

Abstract: Methods, apparatus and systems for wireless sensing and proximity detection are described. For example, a described method comprises: transmitting, by a first wireless device, a wireless signal through a wireless multipath channel in a venue; receiving, by a second wireless device, the wireless signal through the wireless multipath channel, wherein the received wireless signal differs from the transmitted wireless signal due to the wireless multipath channel and a movement of an object in the venue; obtaining a time series of channel information (TSCI) of the wireless multipath channel based on the received wireless signal; computing a first motion statistics based on a time-domain correlation of channel information (CI) in the TSCI; computing a second motion statistics based on a frequency-domain correlation of the CI in the TSCI; and detecting a proximity of the object to a target in the venue based on the first motion statistics and the second motion statistics.

Abstract: According to one embodiment, an estimation device includes a processor. The processor is configured to acquire first and second images including a same subject continuously captured by a capture device installed in a mobile object, acquire a first distance to the subject included in the acquired first image and uncertainty of the first distance, extract a useful second distance from the acquired first distance based on the acquired uncertainty of the first distance, and estimate a position and orientation of the mobile object when the second image is captured by the capture device based on the acquired first and second images, and the extracted second distance.

Abstract: In an image processing method, a detection image and a marked image are obtained. An image segmentation model is applied to segment a first segmented image from the detection image. The first segmented image is corrected according to the marked image to obtain a second segmented image. A size of the second segmented image is adjusted to obtain an adjusted segmented image. The adjusted segmented image is used as a standard segmented image of the detection image. The method improves accuracy of image segmentation and recognition.

Abstract: A method for 2D semantic keypoint detection and tracking is described. The method includes learning embedded descriptors of salient object keypoints detected in previous images according to a descriptor embedding space model. The method also includes predicting, using a shared image encoder backbone, salient object keypoints within a current image of a video stream. The method further includes inferring an object represented by the predicted, salient object keypoints within the current image of the video stream. The method also includes tracking the inferred object by matching embedded descriptors of the predicted, salient object keypoints representing the inferred object within the previous images of the video stream based on the descriptor embedding space model.