Abstract: An image matching apparatus according to the present invention includes: a common region specification unit configured to specify a common region between a first image and a second image; a date replacement unit configured to generate a first replaced image in which a brightness value of the common region of the first image is replaced based on a pixel in the first image, and a second replaced image in which a brightness value of the common region of the second image is replaced based on a pixel in the second image; and a matching unit configured to perform matching between the first image and the second image based on frequency characteristics of the first replaced image and the second replaced image.

Abstract: There are provided methods and computing devices using semi-supervised learning to perform end-to-end video object segmentation, tracking respective object(s) from a single-frame annotation of a reference frame through a video sequence of frames. A known deep learning model may be used to annotate the reference frame to provide ground truth locations and masks for each respective object. A current frame is processed to determine current frame object locations, defining object scoremaps as a normalized cross-correlation between encoded object features of the current frame and encoded object features of a previous frame. Scoremaps for each of more than one previous frame may be defined. An Intersection over Union (IoU) function, responsive to the scoremaps, ranks candidate object proposals defined from the reference frame annotation to associate the respective objects to respective locations in the current frame. Pixel-wise overlap may be removed using a merge function responsive to the scoremaps.

Abstract: A device measures positions of a plurality of objects. The device comprises a memory arranged to store positioning data for a plurality of objects having mutual constant positions, an input interface arranged to obtain measurement data of relative positions for at least a subset of the plurality of objects, and a processor arranged to calculate an adapted measurement position for at least one of the objects of the subset of the plurality of objects by correlating the measured relative positions for the at least a subset of the plurality of objects with the acquired positioning data. The adapted measurement position is based on the positioning data. Corresponding method and computer program are also disclosed.

Abstract: A depth-based object-detection convolutional neural network is disclosed. The depth-based object-detection convolutional neural network described herein incorporates a base network and additional structure. The base network is configured to receive a depth image formatted as RGB image data as input, and compute output data indicative of at least one feature of an object in the RGB image data. The additional structure is configured to receive the output data of the base network as input, and compute predictions of the location of a region in the received depth image that includes the object and of a class of the object as output. An object detection device incorporating the depth-based object-detection convolutional neural network is operable in real time using an embedded GPU.

Abstract: Embodiments of the present disclosure provide a method and apparatus for processing an image, and relates to the field of computer vision technology. The method may include: acquiring a value to be processed, where the value to be processed is associated with an image to be processed; and processing the value to be processed by using a quality scoring model to generate a score of the image to be processed in a target scoring domain, where the score of the image to be processed in the target scoring domain is related to an image quality of the image to be processed.

Abstract: The present invention relates to a method of generating an overhead view image of an area. More particularly, the present invention relates to a method of generating a contextual multi-image based overhead view image of an area using ground map data and field of view image data. Various embodiments of the present technology can include methods, systems and non-transitory computer readable media and computer programs configured to receive a plurality of images of the geographical area, determine a ground map of the geographical area, divide the ground map into a plurality of sampling points of the geographical area; and determine a color for each of the plurality of sampling points, wherein the color of each of the sampling points is determined by determining a correlation between the sampling points of the geographical area and color of the sampling points captured in at least one of the plurality of images.

Abstract: The present disclosure provides a processing device including: a coarse-grained pruning unit configured to perform coarse-grained pruning on a weight of a neural network to obtain a pruned weight, an operation unit configured to train the neural network according to the pruned weight. The coarse-grained pruning unit is specifically configured to select M weights from the weights of the neural network through a sliding window, and when the M weights meet a preset condition, all or part of the M weights may be set to 0. The processing device can reduce the memory access while reducing the amount of computation, thereby obtaining an acceleration ratio and reducing energy consumption.

Abstract: Machine learning multiple features of an item depicted in images. Upon accessing multiple images that depict the item, a neural network is used to machine train on the plurality of images to generate embedding vectors for each of multiple features of the item. For each of multiple features of the item depicted in the images, in each iteration of the machine learning, the embedding vector is converted into a probability vector that represents probabilities that the feature has respective values. That probability vector is then compared with a value vector representing the actual value of that feature in the depicted item, and an error between the two vectors is determined. That error is used to adjust parameters of the neural network used to generate the embedding vector, allowing for the next iteration in the generation of the embedding vectors. These iterative changes continue thereby training the neural network.

Abstract: An electronic apparatus includes a memory configured to store a plurality of images; and a processor configured to identify qualities of the plurality of images, process the plurality of images using at least one artificial intelligence model corresponding to the identified qualities, and obtain a graphic image including the processed plurality of images, and the at least one artificial intelligence model is trained to increase a quality of an input image.

Abstract: Disclosed are an object detection device and a control method. A method for controlling an object detection device comprises the steps of: receiving one image; dividing the received image into a predetermined number of local areas on the basis of the size of a convolutional layer of a convolution neural network (CNN); identifying small objects at the same time by inputting a number of the divided local areas corresponding to the number of CNN channels to each of a plurality of CNN channels; sequentially repeating the identifying of the small objects for each of the remaining divided local areas; selecting MM mode or MB mode; setting an object detection target area corresponding to the number of CNN channels on the basis of the selected mode; and detecting the small objects at the same time by inputting each set object detection target area to each of the plurality of CNN channels.

Abstract: A neural network image processing apparatus arranged to acquire images from an image sensor and to: identify a ROI containing a face region in an image; determine at plurality of facial landmarks in the face region; use the facial landmarks to transform the face region within the ROI into a face region having a given pose; and use transformed landmarks within the transformed face region to identify a pair of eye regions within the transformed face region. Each identified eye region is fed to a respective first and second convolutional neural network, each network configured to produce a respective feature vector. Each feature vector is fed to respective eyelid opening level neural networks to obtain respective measures of eyelid opening for each eye region. The feature vectors are combined and to a gaze angle neural network to generate gaze yaw and pitch values substantially simultaneously with the eyelid opening values.

Abstract: A system employs a trained model to detect artifact(s) associated with artifact type(s) appearing in a reproduction of a source image (a test image). The system determines differences between the test image and the source image and outputs probabilities that the artifact(s) in the test image are associated with each of the artifact type(s). A dataset for training the model includes: (i) a reference category including reference image(s) without any artifacts; and (ii) artifact categories, each corresponding to a respective one of the artifact types and including noised images associated with the respective artifact type. Each noised image includes one of the reference images and an artifact associated with the respective artifact type. The model is trained to detect the artifact type(s) by providing the model with the dataset and causing the model to process differences between each noised image and the reference image in the noised image.

Abstract: A facial recognition system includes a memory for storing a facial image database, wherein the facial image database includes a plurality of entries each corresponding to a different person, and wherein each entry includes a person identifier along with one or more facial images of the person. The facial recognition system further includes a facial recognition module that is operatively coupled to the memory. The facial recognition module is configured to receive a new facial image, and to select one or more facial recognition engines based on one or more facial image parameters of the new facial image, and to use the selected facial recognition engines to compare the new facial image with facial models that are based upon facial images stored in the facial image database in order to identify the person in the new facial image.

Abstract: The present disclosure provides devices, systems and computer-readable media for identifying object characteristics using a machine learning model trained to de-emphasize brightness values. The machine learning model can be trained using modified training data. Modifying the training data can include converting the training data from an original color space into a color space having a brightness channel. The values of the brightness channel for the, training data can then be modified. After the values of the brightness channel are modified, the training data can be converted back into the original color space and used to train the machine learning model. A detection device can be configured with the machine learning model and used to identify object characteristics.

Abstract: When reviewing digital pathology tissue specimens, multiple slides may be created from thin, sequential slices of tissue. These slices may then be prepared with various stains and digitized to generate a Whole Slide Image (WSI). Review of multiple WSIs is challenging because of the lack of homogeneity across the images. In embodiments, to facilitate review, WSIs are aligned with a multi-resolution registration algorithm, normalized for improved processing, annotated by an expert user, and divided into image patches. The image patches may be used to train a Machine Learning model to identify features useful for detection and classification of regions of interest (ROIs) in images. The trained model may be applied to other images to detect and classify ROIs in the other images, which can aid in navigating the WSIs. When the resulting ROIs are presented to the user, the user may easily navigate and provide feedback through a display layer.

Abstract: Disclosed herein includes a system, a method, and a device for improving computational efficiency of deconvolution by reducing a number of dot products. In one aspect, an input image having a set of pixels is received. A first dot product may be performed on a subset of the set of pixels of the input image and a portion of a kernel, to generate a first pixel of an output image. A number of multiplications performed for the first dot product performed may be less than a number of elements of the kernel. A second dot product on a remaining portion of the kernel to generate the first pixel of the output image may be bypassed.

Abstract: A living body detection method and device are disclosed. Wherein the method comprises the following steps: extracting valid depth data of a target detection object from depth map data containing the target detection object; generating a depth difference histogram based on the valid depth data; and inputting the depth difference histogram into a pre-trained machine learning classifier to obtain a determination result of whether the target detection object is a living body. By adopting this method, the detection accuracy can be improved.

Abstract: An intensity transform augmentation system is operable to generate a plurality of sets of augmented images by performing a set of intensity transformation functions on each of a training set of medical scans. Each of the set of intensity transformation functions are based on density properties of corresponding anatomy feature present in the training set of medical scans. A computer vision model is generated by performing a training step on the plurality of sets of augmented images, where each augmented image of a set of augmented images is assigned same output label data based on a corresponding one of the training set of medical scans. Inference data is generated by performing an inference function on a new medical scan by utilizing the computer vision model on the new medical scan. The inference data is transmitted to a client device for display via a display device.

Abstract: Apparatus, systems, and methods to process an interior region of interest in an anatomy image via a user interface are disclosed. An example apparatus is to at least: process an image to reduce noise in the image; identify at least one of an organ of interest or a region of interest in the image; analyze values in at least one of the organ of interest or the region of interest; process the at least one of the organ of interest or the region of interest based on the analyzed values to provide a processed object in the at least one of the organ of interest or the region of interest; and display the processed object for interaction via an interface, the display to include exposing at least one of the organ of interest or the region of interest.

Abstract: A computational photography system is described herein including a guidance system and a detail enhancement system. The guidance system uses a first neural network that maps an original image provided by an image sensor to a guidance image, which represents a color-corrected and lighting-corrected version of the original image. A combination unit combines the original image and the guidance image to produce a combined image. A detail-enhancement system then uses a second neural network to map the combined image to a predicted image. The predicted image supplements the guidance provided by the first neural network by sharpening details in the original image. A training system is also described herein for training the first and second neural networks. The training system alternates in the data it feeds the second neural network, first using a guidance image as input to the second neural network, and then using a corresponding ground-truth image.