Abstract: A fingerprint acquisition device comprising, a space for receiving a finger of a user, a light source configured to emit a illumination light illuminating the space, a camera configured to acquire an image of a fingerprint of the finger, when the finger is received in the space, and, a polarization system configured to polarize the illumination light before the illumination light reaches the space and after the illumination light has passed through the space, and to polarize an ambient light emanating from outside the device and propagating towards the camera, before the ambient light reaches the space, and to polarize the ambient light after the ambient light has passed through the space.

Abstract: A method for providing a global cardiac wall motion classification for a patient is disclosed, and includes employing Cardiac Magnetic Resonance (CMR) image data. In some embodiments, the method comprises one or more of a myocardium segmentation step, a slice classification step, a movement feature extraction step, and a global classification or evaluation step, wherein a patient is classified as having normal cardiac wall motion, or as having suspicious or abnormal cardiac wall motion.

Abstract: Techniques related to performing image registration are discussed. Such techniques include converting a source image region and a target image portion from a color image space to a semantic space and iteratively converging homography parameters using the source image region and target image portion in the semantic space by applying iterations with some homography parameters allowed to vary and others blocked from varying and subsequent iterations with all homography parameters allowed to vary.

Abstract: A method for processing video data, comprising: receiving raw video data, representative of a plurality of frames; detecting, using the raw video data, one or more regions of interest in a detection frame that belongs to the plurality of frames, for example using a region proposal network; performing a cropping process on a portion of the raw video data representative of the detection frame, based on the regions of interest, so as to generate cropped raw video data; performing image processing on the cropped raw video data, including demosaicing, so as to generate processed image data for the detection frame; and analyzing the processed image data, for example using an object detection process, to determine information relating to at least one of said one or more regions of interest.

Abstract: A method includes: acquiring an image of first piece of clothing to be collocated; determining information of one or more second piece of clothing for collocation with the first piece of clothing; determining clothing collocation images containing the information of the one or more second piece of clothing in the collocation image library; and selecting clothing collocation images matched with the image of the first piece of clothing from the determined clothing collocation images. The information of the one or more second piece of clothing is pre-marked clothing category information of clothing collocation images in a collocation image library.

Abstract: An authentication system accesses an image of a face of a user. The face of the user is partially covered by a facial mask. The authentication system detects an area on the facial mask and generates a first identification of the user based on the area on the facial mask. The authentication system also detects an exposed area uncovered by the facial mask on the face of the user and generates a second identification of the user based on the exposed area. The authentication system compares the first identification of the user with the second identification of the user, and authenticates the user based on the comparison.

Abstract: Systems and methods are described for generating pixel image data, using a lensless camera, based on light that travels through a mask that with pattern masking the lensless camera. The system applies a transformation function to the pixel image data to generate frequency domain image data. The system inputs the frequency domain image data into a machine learning model, wherein the machine learning model does not have access to data that represents the pattern of the mask. The model is trained using a set of images with the feature that are captured by the flat, lensless camera through the mask. The system processes the frequency domain image data using the machine learning model to determine whether the pixel image data depicts the image feature. The system further performs an action based on determining that the pixel image data depicts the image feature.

Abstract: An insect sortation system includes a first puff system in a puff-back region for moving insects toward an inlet of an insect pathway and a second puff system in a puff-forward region for moving insects toward an outlet of the pathway. The first puff system is configured to singulate the insects. The second puff system is configured to sort the insects.

Abstract: Accurate vehicle localization is arguably the most critical and fundamental task for autonomous vehicle navigation. While dense 3D point-cloud-based maps enable precise localization, they impose significant storage and transmission burdens when used in city-scale environments. A highly compressed representation for LiDAR maps, along with an efficient and robust real-time alignment algorithm for on-vehicle LiDAR scans, is proposed here. The proposed mapping framework, requires less than 0.1% of the storage space of the original 3D point cloud map. In essence, mapping framework emulates an original map through feature likelihood functions. In particular, the mapping framework models planar, pole and curb features. These three feature classes are long-term stable, distinct and common among vehicular roadways. Multiclass feature points are extracted from LiDAR scans through feature detection.

Abstract: A method of generating a patient-specific prosthetic includes receiving anatomic imaging data representative of a portion of a patient's anatomy. A first digital representation of the anatomic imaging data is defined. The first digital representation of the anatomic imaging data is modified. A second digital representation of the portion of the patient's anatomy is defined based on the modifying of the first digital representation of the anatomic imaging data. A patient-specific prosthetic template of the portion of the patient's anatomy is generated based at least in part on the second digital representation of the anatomic imaging data.

Abstract: A sensor device includes an array sensor having a plurality of detection elements arrayed in one or two dimensional manner, a signal processing unit configured to acquire a detection signal by the array sensor and perform signal processing, and a calculation unit. The calculation unit detects an object from the detection signal by the array sensor, and gives an instruction, to the signal processing unit, on region information generated on the basis of the detection of the object as region information regarding the acquisition of the detection signal from the array sensor or the signal processing for the detection signal.

Abstract: An electronic device is provided that includes a display in which a fingerprint recognition area is formed and which includes a display pixel layer, and a rear panel, formed under the display, having a first hole. The electronic device also includes a bracket, disposed under the rear panel, having a sensor accommodating area corresponding to a fingerprint sensor. The electronic device additionally includes the fingerprint sensor disposed under the display pixel layer and adapted to acquire image information to be used for authentication of a fingerprint using at least partially light radiated from the display pixel layer and reflected by the object. The sensor accommodating area has a second hole passing through the bracket, and the fingerprint sensor is disposed in the sensor accommodating area such that an air gap between the display and the fingerprint sensor is formed at least by a portion of the second hole.

Abstract: A method comprises receiving one or more inputs captured by a camera of a device, and determining, using one or more machine learning models, whether the one or more inputs depict at least one object configured to capture a visual representation of a screen of the device. A recommendation is generated responsive to an affirmative determination, the recommendation comprising at least one action to prevent the capture of the visual representation of the screen of the device.

Abstract: Image captured through a non-rectilinear lens may exhibit distortions. The distortions may be reduced by warping the image. However, warping the image may degrade the fidelity of the image. The warped image may be enhanced to increase the fidelity of the image. The enhancement may be applied to the portions of the image that were degraded from the warping.

Abstract: The present invention provides an automatic container loading and unloading apparatus and method. The apparatus comprises: a data acquisition module, used for scanning a container truck panel to obtain laser point cloud data; a data preprocessing module, used for segmenting a laser point cloud on a surface of the container truck panel from the laser point cloud data; a key point extraction module, used for performing edge extraction on the laser point cloud on the surface of the container truck panel to obtain discrete points on edges of the keel of the container truck panel; and a straight line fitting module, used for performing random sample consensus straight line fitting on the discrete points on the edges of the keel of the container truck panel to obtain spatial straight lines of the edges of the keel of the truck panel.

Abstract: A display apparatus with fingerprint identification module within the display region of the display apparatus comprises a first cover and a display module, the fingerprint identification module being between the two. The first cover defines a display region and a non-display region. The fingerprint identification module includes a first substrate, a touch unit, a fingerprint identification unit, a control unit, and a first dummy unit with a plurality of first dummy electrodes. The first dummy unit is coplanar with the fingerprint identification unit. The first dummy electrodes surround the fingerprint identification unit, and cooperate with the fingerprint identification unit to enable uniform light transmittance over the whole display region.

Abstract: Provided is an image restoration method of restoring an image obtained through an array camera, the image restoration method including obtaining a plurality of images through lens elements included in the array camera, obtaining a global parameter of the plurality of images, generating first processed images by transforming a viewpoint of each of the plurality of images based on the obtained global parameter, obtaining a local parameter for each pixel corresponding to each of the first processed images, generating second processed images by transforming the first processed images based on the obtained local parameter, and generating a synthesized image of a target viewpoint based on synthesizing the second processed images.

Abstract: Disclosed herein are methods and systems for performing instance segmentation that can provide improved estimation of object boundaries. Implementations can include a machine-learned segmentation model trained to estimate an initial object boundary based on a truncated signed distance function (TSDF) generated by the model. The model can also generate outputs for optimizing the TSDF over a series of iterations to produce a final TSDF that can be used to determine the segmentation mask.

Abstract: Images of a hand are obtained by a camera. These images may depict the fingers and palm of the user. A pose of the hand relative to the camera may vary due to rotation, translation, articulation of joints in the hand, and so forth. One or more canonical images are generated by mapping the images to a canonical model. A first embedding model is used to determine a first embedding vector representative of the palm as depicted in the canonical images. A second embedding model is used to determine a set of second embedding vectors, each representative of individual fingers as depicted in the canonical images. Embedding distances in the embedding space from the embedding vectors to a closest match of previously stored embedding vectors are multiplied together to determine an overall distance. If the overall distance is less than a threshold value, an identity of a user is asserted.

Abstract: A method for processing an image by using a neural network is provided. The method includes obtaining a target image; and obtaining a restored image corresponding to the target image by using the neural network, the neural network including a plurality of residual blocks connected sequentially. The obtaining the restored image includes obtaining the restored image by performing, in each residual block of the plurality of residual blocks, image processing on an input image, which is input from a previous residual block, on different scales, and providing results of the image processing performed on the different scales to a subsequent residual block.