Abstract: A moving body speed derivation method that causes at least one computer to execute a process, the process includes deriving a reference point in a second moving body region that is an image of a second moving body, from one image capturing the second moving body among a plurality of time-series images captured by an imaging device installed in a first moving body; deriving a position of the second moving body based on a distance between the imaging device and a feature point in the second moving body corresponding to the reference point in the second moving body region; and deriving speed of the second moving body based on a change amount of the position of the second moving body corresponding to the second moving body region included in the plurality of time-series images.

Abstract: A security camera with a motion detection function, which comprises: an image sensor, configured to capture original images; a variation level computation circuit, configured to compute image variation levels of the original images; a long term computation circuit, configured to calculate a first average level for the image variation levels corresponding to M of the original images; a short term computation circuit, configured to calculate a second average level for the image variation levels corresponding to N of the original images, wherein M>N; and a motion determining circuit, configured to determine whether a motion of an object appears in a detection range of the image sensor according to a relation between the first average level and the second average level. By such security camera, the interference caused by noise or small object can be avoided. Accordingly, the motion detection of the security camera can be more accurate.

Abstract: Methods and image processing systems are provided for determining a dominant gradient orientation for a target region within an image. A plurality of gradient samples are determined for the target region, wherein each of the gradient samples represents a variation in pixel values within the target region. The gradient samples are converted into double-angle gradient vectors, and the double-angle gradient vectors are combined so as to determine a dominant gradient orientation for the target region.

Abstract: Systems and methods are described for display of a depth image (depth plus texture) using multiple focal planes. In one embodiment, a depth image (which may be a frame of a depth video, consisting of a video plus depth sequence) is mapped to a first set of image planes. The depth image (or a subsequent frame of the depth video) is mapped to a second set of image planes. Each image plane in the first and second set has a specified depth, and the first and second set differ in at least one depth. Each of the image planes is displayed in the first set at the respective depth of that image plane, and, subsequently, each of the image planes in the second set is displayed at its respective depth. Display of the first and second sets may be cyclically alternated at rate sufficiently high to avoid perceptible flicker.

Abstract: Electronic device depth camera systems having a reduced number of outwardly-facing optical components are disclosed. Monocular depth camera systems, including basic depth camera systems as well as RGB-D camera systems, have exactly one window on a housing panel of the device through which light is transmittable out of and into the device housing. An optical emitter and detector(s) are located within the device housing. Binocular RGB-D camera systems have exactly two outwardly-facing optical components on the housing panel. One of the components, such as an optical emitter or an exit window, is associated with illumination light leaving the device, and the other optical component is an entrance window associated with transmitting detection light into the housing.

Abstract: An image sensor is positioned such that a field-of-view of the sensor encompasses portion of a beverage machine. The field-of-view includes a zone associated with operating the beverage machine. A subsystem receives a signal indicating beverage is dispensed and angled-view images from the image sensor. First image(s) associated with a start of beverage being dispensed are determined by detecting a hand of a person entering the zone associated with operating the beverage machine. A first identifier is determined for the person in the first image(s). Second image(s) associated with an end of the beverage being dispensed are determined by detecting a hand of the person exiting the zone. A second identifier is determined for the person in the second image(s). If the first identifier is the same as the second identifier, the beverage is assigned to the person with the first identifier.

Abstract: A color calibration system for color correction in an optical device using a database of target colors includes a physical target card. The physical target card includes at least three unique colored patches arranged in a predetermined pattern on the physical target card, identifying indicia, and alignment indicia. The color calibration system also includes an executable machine-readable software. The software is configured to read and assign a value to the colored patches. The software is further configured to read the identifying indicia and identify the physical target card, and to read the alignment indicia for identifying individual colored patches. The software is further configured to compare color values in the optical device to known factory color values, generate a target card profile, and convert native color space for use in subsequent images under the same lighting conditions.

Abstract: An object tracking system that includes a first sensor and a second sensor that are each configured to capture frames of at least a portion of a global plane for a space. The system is configured to identify a first pixel location for a marker within a first frame and to determine an (x,y) coordinate for the marker using a first homography. The system is further configured to identify a second pixel location for the marker in the second sensor using a second homography, to identify a third pixel location using a disparity mapping, and to determine a distance difference between the second pixel location and the third pixel location. The system is further configured to compare the distance difference to a difference threshold level and to recompute the first homography and/or the second homography in response to determining that the distance difference exceeds the difference threshold level.

Abstract: Disclosed is a compression system for compressing image data. The compression system receives an uncompressed image file with data points that are defined with absolute values for elements representing the data point position in a space. The compression system stores the absolute values defined for a first data point in a compressed image file, determines a difference between the absolute values of the first data point and the absolute values of a second data point, derives a relative value for the absolute values of the second data point from the difference, and stores the relative value in place of the absolute values of the second data point in the compressed image file.

Abstract: An image processing apparatus including a decoder configured to identify an area in an image frame at which object recognition is to be performed by using type information of a macroblock included in the image frame, and perform object recognition on the area of the image frame and skip the object recognition on areas of the image frame other than the area.

Abstract: A method for lossy image or video encoding, transmission and decoding, the method comprising the steps of: receiving an input image at a first computer system; encoding the input image using a first trained neural network to produce a latent representation; identifying one or more regions of the input image associated with high visual sensitivity; encoding the one or more regions of the input image associated with high visual sensitivity using a second trained neural network to produce one or more region latent representations; performing a quantization process on the latent representation and the one or more region latent representations; transmitting the result of the quantization process to a second computer system; decoding the result of the quantization process to produce an output image, wherein the output image is an approximation of the input image.

Abstract: The present disclosure relates to systems, methods, and non-transitory computer readable media for generating a modified digital image from extracted spatial and global codes. For example, the disclosed systems can utilize a global and spatial autoencoder to extract spatial codes and global codes from digital images. The disclosed systems can further utilize the global and spatial autoencoder to generate a modified digital image by combining extracted spatial and global codes in various ways for various applications such as style swapping, style blending, and attribute editing.

Abstract: An angled optical pattern is decoded. To decode an optical pattern imaged at an angle, an area of interest of an image is received. A start line and an end line of the optical pattern are estimated. Corners of the optical pattern are localized. A homography is calculated based on the corners. And a scanline of the optical pattern is rectified based on the homography.

Abstract: A method and apparatus for recognizing an imaged information-bearing medium, a computer-readable storage device and a computer device are provided. The method comprising: acquiring a first image of the imaged information-bearing medium; performing text recognition on the first image to acquire a text content of the imaged information-bearing medium; classifying the imaged information-bearing medium to acquire a type of the imaged information-bearing medium; and archiving the text content according to the type.

Abstract: A neural network system, includes: a processor configured to detect a plurality of object candidates included in a first image, generate metadata corresponding to the plurality of object candidates based on the first image, and set data processing orders of the plurality of object candidates based on the metadata; and at least one resource configured to perform data processing with respect to the plurality of object candidates. The processor is configured to sequentially provide pieces of information related to data processing of the plurality of object candidates to the at least one resource according to the set data processing orders, and the at least one resource is configured to sequentially perform data processing with respect to the plurality of object candidates according to an order in which a piece of information related to data processing of each of the plurality of object candidates is received.

Abstract: A text recognition method and apparatus disclosed. The text recognition method includes: obtaining a to-be-detected image; determining a target text detection area in the to-be-detected image, where the target text detection area includes target text in the to-be-detected image, and the target text detection area is a polygonal area including m vertex pairs, m being a positive integer greater than 2; correcting the polygonal area to m?1 rectangular areas to obtain a corrected target text detection area; and performing text recognition on the corrected target text detection area to determine the target text, and outputting the target text.

Abstract: The present invention discloses a method and system for recognizing a geometric regularity image of a honeycomb structure. The method includes the steps of image acquisition, image processing, vertex extraction, cell reconstruction, and quality evaluation, wherein a step of binaryzation is set between the step of image processing and the step of vertex extraction, and is to set a pixel value of a background in the image to be 0 and set a pixel value of a honeycomb skeleton in the image to be 1 to form a binary image, and the step of quality evaluation is to calculate angular deviation values of all the cells and an average thereof as well as linear deviation values and an average thereof based on the reconstructed cell image, and determine whether the honeycomb structure is qualified or not by comparing with a set tolerance zone.

Abstract: A computer-implemented method of generating metadata from an image may comprise sending the image to an object detection service, which generates detections metadata from the image. The image may also be sent to a visual features extractor, which extracts visual features metadata from the image. The generated detections metadata may then be sent to an uncertainty score calculator, which computes an uncertainty score from the detections metadata. The uncertainty score may be related to a level of uncertainty within the detections metadata. The image, the visual features metadata, the detections metadata and the uncertainty score may then be stored in a database accessible over a computer network.

Abstract: Provided are a method for converting a picture into a video, a device, and a storage medium. The method for converting a picture into a video includes the following: performing color fading processing on an original picture to obtain a black-and-white image; determining a foreground area and a background area of the black-and-white image; performing iterative color restoration on pixels in the black-and-white image based on a processing order of the foreground area and the background area, and storing the image obtained by each restoration as a picture frame to obtain multiple frames of images; and splicing the multiple frames of images to obtain a target video.

Abstract: Provided is an image acquisition apparatus including a first hyperspectral image sensor configured to obtain a first image, and a processor configured to process the first image obtained by the first hyperspectral image sensor, wherein the processor is further configured to extract a spectrum of each pixel of the first image obtained by the first hyperspectral image sensor, correct the spectrum by normalizing the spectrum of each pixel, distinguish an object from a background within the first image by recognizing a pattern of the spectrum of each pixel, and correct a color of the object and a color of the background by using a spatial distribution of illuminations.

Abstract: Learning iterations, individual ones of which include a respective bucket group selection phase and a class boundary refinement phase, are performed using a source data set whose records are divided into buckets. In the bucket group selection phase of an iteration, a bucket is selected for annotation based on output obtained from a classification model trained in the class boundary refinement phase of an earlier iteration. In the class boundary refinement phase, records of buckets annotated as positive-match buckets for a target class in the bucket group selection phase are selected for inclusion in a training set for a new version of the model using a model enhancement criterion. The trained version of the model is stored.

Abstract: A finger vein comparison method, a computer equipment, and a storage medium are provided. The finger vein comparison method includes: two finger vein images to be compared are obtained (S10); image channel fusion is performed on the two finger vein images to be compared to obtain a two-channel target finger vein image to be compared (S20); the target finger vein image to be compared is input into a feature extractor, and a feature vector of the target finger vein image to be compared is extracted by the feature extractor (S30); the feature vector of the target finger vein image to be compared is input into a dichotomy classifier to obtain a dichotomy result (S40); and it is determined according to the dichotomy result whether the two finger vein images to be compared come from the same finger (S50).

Abstract: Systems and methods for training machine models with augmented data. An example method includes identifying a set of images captured by a set of cameras while affixed to one or more image collection systems. For each image in the set of images, a training output for the image is identified. For one or more images in the set of images, an augmented image for a set of augmented images is generated. Generating an augmented image includes modifying the image with an image manipulation function that maintains camera properties of the image. The augmented training image is associated with the training output of the image. A set of parameters of the predictive computer model are trained to predict the training output based on an image training set including the images and the set of augmented images.

Abstract: A user-generated graphical representation can be sent into a generative network to generate a synthetic image of an area including a road, the user-generated graphical representation including at least three different colors and each color from the at least three different colors representing a feature from a plurality of features. A determination can be made that a discrimination network fails to distinguish between the synthetic image and a sensor detected image. The synthetic image can be sent, in response to determining that the discrimination network fails to distinguish between the synthetic image and the sensor-detected image, into an object detector to generate a non-user-generated graphical representation. An objective function can be determined based on a comparison between the user-generated graphical representation and the non-user-generated graphical representation.

Abstract: An image recognition method and apparatus, a training method, an electronic device, and a storage medium are provided. The image recognition method includes: acquiring an image to be recognized, the image to be recognized including a target text; and determining text content of the target text based on knowledge information and image information of the image to be recognized.

Abstract: Systems, methods, and computer program products for image recognition in which instructions are executable by a processor to dynamically generate simulated documents and corresponding images, which are then used to train a fully convolutional neural network. A plurality of document components are provided, and the processor selects subsets of the document components. The document components in each subset are used to dynamically generate a corresponding simulated document and a simulated document image. The convolutional neural network processes the simulated document image to produce a recognition output. Information corresponding to the document components from which the image was generated is used as an expected output. The recognition output and expected output are compared, and weights of the convolutional neural network are adjusted based on the differences between them.

Abstract: A method of controlling operation of equipment that spools an elongated member on and off a rotatable drum, in one or more embodiments, includes obtaining video data of a position of the elongated member on the rotatable drum. The method can also include feeding data into a trained artificial neural network and processing the data fed into the trained artificial neural network to determine at least one of a calculated position of the elongated member on the rotatable drum, a calculated fleet angle, or both. The method can also include actuating the rotatable drum to one of spool elongated member on and off the rotatable drum.

Abstract: Systems and methods for detecting motile objects (e.g., parasites) in a fluid sample by utilizing the locomotion of the parasites as a specific biomarker and endogenous contrast mechanism. The imaging platform includes one or more substantially optically transparent sample holders. The imaging platform has a moveable scanning head containing light sources and corresponding image sensor(s) associated with the light source(s). The light source(s) are directed at a respective sample holder containing a sample and the respective image sensor(s) are positioned below a respective sample holder to capture time-varying holographic speckle patterns of the sample contained in the sample holder. The image sensor(s). A computing device is configured to receive time-varying holographic speckle pattern image sequences obtained by the image sensor(s). The computing device generates a 3D contrast map of motile objects within the sample use deep learning-based classifier software to identify the motile objects.

Abstract: The present disclosure provides a method for image segmentation, an electronic device and a computer readable storage medium, relating to the field of computer vision technology and artificial intelligence technology. The method comprises: performing feature extraction on an image through a feature extraction network to obtain a feature map; performing feature map processing on the feature map to complete semantic segmentation of the image. The disclosure may effectively improve performance of semantic segmentation.

Abstract: Certain aspects involve a machine-learning query system that uses a dual deep learning network to service queries and other requests. In one example, a machine-learning query system services a query received from a client computing system. A dual deep learning network included in the machine-learning query system matches an unstructured input data object, received from the client computing system, to an unstructured reference data object. The matching may include generating an input feature vector by an embedding subnetwork, based on the unstructured input data object. The matching may also include generating an output probability by a relationship subnetwork, based on the input feature vector and a relationship feature vector that is based on the unstructured reference data object. The machine-learning query system may transmit a responsive message to the client system.

Abstract: In one aspect, an example method includes a processor (1) applying a feature map network to an image to create a feature map comprising a grid of vectors characterizing at least one feature in the image and (2) applying a probability map network to the feature map to create a probability map assigning a probability to the at least one feature in the image, where the assigned probability corresponds to a likelihood that the at least one feature is an overlay. The method further includes the processor determining that the probability exceeds a threshold, and responsive to the processor determining that the probability exceeds the threshold, performing a processing action associated with the at least one feature.

Abstract: A neural processing unit (NPU) for decoding video or feature map is provided. The NPU may comprise at least one processing element (PE) to perform an inference using an artificial neural network. The at least one PE may be configured to receive and decode data included in a bitstream. The data included in the bitstream may comprise data of a base layer. Alternatively, the data included in the bitstream may comprise data of the base layer and data of at least one enhancement layer. The data of the base layer included in the bitstream may include a first feature map. The data of the at least one enhancement layer included in the bitstream may include a second feature map.

Abstract: Aspects of the disclosure provide for systems and processes for assessing image quality for optical character recognition (OCR), including but not limited to: segmenting an image into patches, providing the segmented image as an input into a first machine learning model (MLM), obtaining, using the first MLM, for each patch, first feature vectors representative of a reduction of imaging quality in a respective patch, and second feature vectors representative of a text content of the respective patch, providing to a second MLM the first feature vectors and the second feature vectors, and obtaining, using the second MLM, an indication of suitability of the image for OCR.

Abstract: Embodiments of this application disclose an object annotation method and apparatus, a movement control method and apparatus, a device, and a storage medium. The method includes: obtaining a reference image recorded by an image sensor from an environment space, the reference image comprising at least one reference object; obtaining target point cloud data obtained by a three-dimensional space sensor by scanning the environment space, the target point cloud data indicating a three-dimensional space region occupied by a target object in the environment space; determining a target reference object corresponding to the target object from the reference image; determining a projection size of the three-dimensional space region corresponding to the target point cloud data and the three-dimensional space region being projected onto the reference image; and performing three-dimensional annotation on the target reference object in the reference image according to the determined projection size.

Abstract: Systems and techniques for performing a scan-free reading of a structured identifier are provided. One technique involves obtaining a first set of images of an object within a scanning volume of an environment. The object is passed through the scanning volume without an explicit presentation of the structured identifier to an optical scanner within the environment. For each image of a plurality of the first set of images, an indication of a first region of interest of the structured identifier on the object within the image is obtained, based in part on a second set of images of the object having a lower quality than the first set of images. The first region of interest of the structured identifier is extracted from a first image of the plurality of the first set of images. The structured identifier within the first region of interest within the first image is decoded.

Abstract: Disclosed is a system including a content receiver and an antenna coupled to the content receiver. A camera is mounted on the antenna. The content receiver includes a processor and a memory storing instructions that, when executed by the processor, cause the content receiver to: output a control signal to the camera mounted on the antenna, receive image data from the camera mounted on the antenna, determine that the image data received from the camera mounted on the antenna indicates a potential or actual obstruction of the antenna or movement of the antenna, and output a message indicating that the potential or actual obstruction of the antenna or movement of the antenna device has been detected, in response to determining that the image data received from the camera mounted on the antenna indicates the potential or actual obstruction of the antenna or movement of the antenna.

Abstract: Stubble burning is a serious problem resulting in pollution attributable to smog, loss of nutrients in the top soil, and risk of fires spreading out of control. Existing methodologies have attempted to predict burning areas, but have failed to do so because of inefficient sensing mechanism. Present disclosure proposes a system and method to compute burning index score pertaining to crops by detecting harvest season and predicting probable areas of burning by combining current year's crop area map along with harvesting period and historical hot spot information. Computation of the burning index score is accomplished based on inputs received from at least one of satellite imaging, multi-spectral drone based sensing devices and crowdsourcing information. This will help to prioritize the area for taking corrective measures such as training of farmers, availability of resources, optimizing the resources schedule, etc.

Abstract: A deep neural network provides real-time pose estimation by combining two custom deep neural networks, a location classifier and an ID classifier, with a pose estimation algorithm to achieve a 6D0F location of a fiducial marker. The locations may be further refined into subpixel coordinates using another deep neural network. The networks may be trained using a combination of auto-labeled videos of the target marker, synthetic subpixel corner data, and/or extreme data augmentation. The deep neural network provides improved pose estimations particularly in challenging low-light, high-motion, and/or high-blur scenarios.

Abstract: Systems and methods are provided for performing operations including: receiving, via a messaging application, input that selects a collection of augmented reality items; obtaining an identifier of the collection of the augmented reality items; searching, based on the identifier, a plurality of augmented reality items to identify a subset of augmented reality items associated with the identifier; causing the messaging application to present the subset of augmented reality items; and causing the messaging application to modify an image based on a first augmented reality item in the subset.

Abstract: The present disclosure relates to systems, methods, and computer-readable media for selectively identifying image frames from an input video to provide to an image processing model based on camera statistics. For example, systems disclosed herein include receiving an input video and associated camera statistics from a video capturing device. The systems disclosed herein further include identifying select image frames to provide to the image processing model based on the camera statistics and based on an application of the image processing model. The systems disclosed herein further include selectively identifying and providing camera statistics to the image processing model. By selectively providing data to the image processing model based on camera statistics, the systems disclosed herein can leverage capabilities of video capturing devices to significantly reduce the expense of processing resources when utilizing a variety of image processing models.

Abstract: Techniques are disclosed for identifying and presenting video content that demonstrates features of a target product. The video content can be accessed, for example, from a media database of user-generated videos that demonstrate one or more features of the target product so that a user can see and hear the product in operation via a product webpage before making a purchasing decision. The product functioning videos supplement any static images of the target product and the textual product description to provide the user with additional context for each of the product's features, depending on the textual product description. The user can quickly and easily interact with the product webpage to access and playback the product functioning video to see and/or hear the product in operation.

Abstract: A plurality of on-road vehicles moving along streets of a city, while utilizing onboard cameras for autonomous driving functions, are bound to capture, over time, random images of various city objects such as pedestrians, vehicles, building and other structures. Images of a certain city object may be captured by many of the vehicles at various times while passing by that object, thereby resulting in a corpus of imagery data containing various random images and other representations of that certain object. Per each of at least some of the city objects appearing more than once in the corpus of imagery data, appearances are detected and linked, and then analyzed to reveal behavioral/descriptive models associated with that city object and with interactions of the object with other objects. The models are then used to detect a recent emerging pattern that is then used to determine an appropriate service action.

Abstract: Embodiments are directed to segmentation and hierarchical clustering of video. In an example implementation, a video is ingested to generate a multi-level hierarchical segmentation of the video. In some embodiments, the finest level identifies a smallest interaction unit of the video—semantically defined video segments of unequal duration called clip atoms. Clip atom boundaries are detected in various ways. For example, speech boundaries are detected from audio of the video, and scene boundaries are detected from video frames of the video. The detected boundaries are used to define the clip atoms, which are hierarchically clustered to form a multi-level hierarchical representation of the video. In some cases, the hierarchical segmentation identifies a static, pre-computed, hierarchical set of video segments, where each level of the hierarchical segmentation identifies a complete set (i.e., covering the entire range of the video) of disjoint (i.e.

Abstract: In aspects, a visitor interaction system is described for interacting with visitors. The visitor interaction system can obtain video data, identify an approach of a person, and determine an identity of the person. Further, the visitor interaction system can determine that the person belongs to a respective visitor group of a plurality of visitor groups. Each of the plurality of visitor groups corresponds to a response model that includes a plurality of response actions. At least one of the response actions includes an autonomous response action. The visitor interaction system can also identify a response model corresponding to the respective visitor group and initiate an autonomous response action associated with the response model. The visitor interaction system can then send a report message to the user via a client device.

Abstract: A computer-implemented method, comprising: obtaining motion indicators for a plurality of samples of a video stream; obtaining an anomaly state for each of a plurality of time windows of the video stream, each of the time windows spanning a subset of the samples, by (i) obtaining estimated statistical parameters for the given time window based on measured statistical parameters characterizing the motion indicators for the samples in at least one time window of the video stream that precedes the given time window and (ii) determining the anomaly state for the given time window based on the plurality of motion indicators obtained for the samples in the given time window and the estimated statistical parameters; and processing the video stream based on the anomaly state for various ones of the time windows.

Abstract: An information processing device of the present invention includes a detection means that detects the content of an image, a determination means that determines a processing mode for the image based on the result of detection of the content of the image, and an execution means that executes processing for a captured image corresponding to the processing mode.

Abstract: Methods and systems for crowd level estimation are provided. The system for crowd level estimation includes an input module and a crowd level estimation module. The input module receives an input image of a crowd and the crowd level estimation module estimates a crowd level of the crowd in the input image in response to a crowd density level.

Abstract: Technologies and techniques for vehicle perception. Contours are extracted from image data, and a first and second contour of a current image and next image are determined along a plurality of image row boundaries, wherein the contours includes a plurality of contour points linked to each other via respective edges. An image velocity vector is applied to project each of the plurality of first contour points to the next image, wherein at least some of the projected first contour points are not located on an image row boundary. The projected contour points are processed to determine projected edges forming a projected link between each projected contour point. The projected contour points are moved along the projected link to align each of the moved projected contour points to a respective image boundary. Each of the moved projected contour points are then linked to each of the second contour points.

Abstract: A scene reproduction test method, apparatus, device and program product of an autonomous driving system are provided by the present application. The method includes: obtaining problematic scene data generated during a test on an autonomous driving vehicle equipped with a first autonomous driving system, and determining key obstacle information and non-key obstacle information according to the problematic scene data; generating, according to the non-key obstacle information, a virtual obstacle model, and determining, according to the key obstacle information, a real device used to simulate a key obstacle; where the real device is a device that actually exists in a test environment; performing, by utilizing the virtual obstacle model and the real device, a reproduction test on an autonomous driving vehicle equipped with a second autonomous driving system. In the solution, the test object is a real autonomous driving vehicle, and the key obstacle is a real device.

Abstract: A vehicle and a system and method of operating the vehicle based on a gesture made by a traffic director. The system includes a camera and at least one neural network. The camera obtains an image of a flag operator. The at least one neural network is to generates an encoded hand vector based on a configuration of a hand of the traffic director from the image, combines a skeleton of the traffic director generated from the image and the encoded hand vector to generate a representation vector, and predicts a gesture of the traffic director from the representation vector.