Abstract: Provided is a data labeling method based on artificial intelligence, an apparatus, and a storage medium relating to the field of artificial intelligence, particularly data labeling, image recognition, and natural language processing. The method includes: determining a plurality of samples involved in clustering; performing a plurality of following operations circularly to realize iterative processing, until a convergence condition is satisfied or a quantity of iterations reaches a number threshold, comprising: pre-clustering the plurality of samples according to a vector representation of the respective samples to obtain a plurality of class clusters, each class cluster containing at least one sample; receiving labeling information for the respective class clusters and re-determining the plurality of samples according to the labeling information; and determining a clustering result according to the labeling information for the respective class clusters.

Abstract: An automated camera system and method are provided that are configured to search for an object emitting or reflecting a certain wavelength of light. The camera can be infrared, electro-optical, or otherwise. Image processing software is configured to isolate pixel groups in each image, wherein those pixel groups fit a wavelength or color patterns of an object being searched. The system also includes processing methods for determining the relative location of the object (GPS or other) based upon the location of the camera and the field of view of the camera. Materials can also be used to tag the object so that the object is easier to find in adverse lighting conditions. Alternatively, the tag can be a wavelength that only one group of searchers knows about, and hence the object can remain relatively hidden from others. The system can be readily mounted in an aircraft or other vehicle so that large areas can be scanned and searched.

Abstract: A gaze point calculation apparatus includes: a first cache register, a multiplexer, an arithmetic unit assembly, and a state machine. The first cache register is configured to receive and store first coordinates and a plurality of calibration parameters required when second required are obtained through calculation according to the first coordinates. The state machine is configured to control the multiplexer to select each time at least one value from the first cache register and transmit same to the arithmetic unit assembly. The arithmetic unit assembly is configured to perform a preset operation on the at least one value received each time until the second coordinates are obtained, and output the second coordinates under control of the state machine.

Abstract: Embodiments are disclosed for generating an instant mask from polarized input images.

Abstract: Disclosed is a saturation enhancement method, the method comprising: performing color space conversion on an input image to obtain color data of color conversion; performing saturation expansion on the color data of the color conversion to obtain expanded color data; and performing Gamma preprocessing on the expanded color data to obtain processed color data. Also disclosed are a saturation enhancement device and a computer readable storage medium.

Abstract: A method of complementing a map of a scene with 3D reference points including four steps. In a first step, data is collected and recorded based on samples of at least one of an optical sensor, a GNSS, and an IMU. A second step includes initial pose generation by processing of the collected sensor data to provide a track of vehicle poses. A pose is based on a specific data set, on at least one data set recoded before that dataset and on at least one data set recorded after that data set. A third step includes SLAM processing of the initial poses and collected optical sensor data to generate keyframes with feature points. In a fourth step 3D reference points are generated by fusion and optimization of the feature points by using future and past feature points together with a feature point at a point of processing. This second and fourth steps provides significantly better results than SLAM or VIO methods known from prior art, as the second and the fourth steps are based on recorded data.

Abstract: An image processing system includes one or more memories; and one or more processors coupled to the one or more memories and the one or more processors configured to acquire a degree of influence of each region of image data on a recognition result when recognition processing is performed for the image data, and reduce data size of the image data based on the degree of influence.

Abstract: A method for tracking occluded objects includes encoding locations of a plurality of objects in an environment, determining a target object, receiving a first end point corresponding to a position of the target object before occlusion behind an occlusion object, distributing a hypothesis between both sides of the occlusion object during occlusion from a subsequent frame of the sequence of frames, receiving a second end point corresponding to a position of the target object after emerging from occlusion from another subsequent frame of the sequence of frames, and determining a trajectory of the target object when occluded by the occlusion object by performing inferences using a spatio-temporal probabilistic graph based on the current frame and the subsequent frames of the sequence of frames. The trajectory of the target object when occluded is used as a learning model for future target objects that are occluded by the occlusion object.

Abstract: The present invention addresses the problem of providing an image processing system that can accurately and efficiently detect a target object even when the positional relationship between an image capturing device and the target object differs between the time of teaching and the time of detection.

Abstract: A system is provide that is configured to execute a contest that includes displaying an image of a face of a model, among a plurality of models, to a user; receive inputs from the user to adjust a color of a blended cosmetic product to be applied to the face of the model; display the model with the adjusted color selected by the user; receive the image of the face of the model with the adjusted color as an entry of the user; display the entry of the user to a plurality of users and provide the plurality of users an option submit a vote on the entry of the users and entries of other users; output a result of a contest based on one or more receiving a highest number of votes. The system further transmits an instruction to a manufacturing facility of cartridges of s dispensing device to create a grouped package of cartridges for the determined plurality of color ingredients.

Abstract: In one embodiment, a method includes accessing an input image associated with an initial illumination, extracting from the input image at least a first, a second, and a third mono-channel images based on a first, a second, a third image channel, respectively, determining a first, a second, and a third correction for the first, second, and third mono-channel images with a first, a second, and a third correction algorithm, respectively, generating a first, a second, and a third corrected mono-channel image by applying the first, second, and third correction to the first, second, and third mono-channel image, respectively, and generating an output corrected image corresponding to the input image based on the first, second, and third mono-channel corrected images, wherein the output corrected image is associated with a corrected illumination with respect to the initial illumination.

Abstract: A non-contact facial blood pressure measurement method based on 3D CNN is disclosed, which belongs to the technical field of computer vision. The method includes the following steps. S110: collecting an actual face video sample and training a blood pressure prediction model based on face images using 3D CNN neural network. S120: obtaining a face video in real time through a HD camera. S130: recognizing face key points in the face video obtained in S120 through dlib face recognition model, selecting a face region of interest, and extracting face images from the region. S140: performing a wavelet transform operation on the face images extracted in S130 to remove noise. S150: inputting seven consecutive frames of the face images into the 3D CNN blood pressure prediction model trained in S110 to obtain a blood pressure value of the measured person. The disclosure realizes non-contact facial blood pressure measurement.

Abstract: Systems, methods, and software of inspecting growth areas for plants. In one embodiment, an inspection system captures a plurality of digital images of the growth area from different angles in relation to the growth area, and processes the digital images to identify a boundary of the growth area in the digital images. The inspection system combines the digital images based on the boundary identified in the digital images to generate a composite image of the growth area, and performs image processing on the composite image to detect one or more deficient growth regions in the growth area. The inspection system highlights the deficient growth regions in the composite image, and displays the composite image with the deficient growth regions highlighted.

Abstract: Method and systems are described that create a 3D reconstruction of a vessel of interest that represents a subset of a coronary tree that includes a lesion; calculate at least one of pressure parameters or anatomical parameters based at least in part on a portion of the 3D reconstruction that includes the lesion; calculate a wall shear stress (WSS) descriptor, based on the 3D reconstruction, for a segment of a surface of the vessel that includes the lesion, wherein the WSS descriptor includes information regarding an amount of variation in contraction or expansion applied at surface elements within the segment during at least a portion of a cardiac cycle; and calculate a myocardial infarction (MI) index based on the WSS descriptor and the at least one of the pressure or anatomical parameters, the MI index representing a likelihood that the lesion will result in an MI.

Abstract: Disclosed are a high-precision semi-automatic image data labeling method, an electronic apparatus and a non-transitory computer-readable storage medium. The high-precision semi-automatic image data labeling method may include: displaying a to-be-labeled image, the to-be-labeled image comprising a selected area and an unselected area; acquiring a coordinate point of the unselected area and a first range value; executing a grabcut algorithm based on the coordinate point of the unselected area and the first range value acquired, and obtaining a binarized image divided by the grabcut algorithm; executing an edge tracking algorithm on the binarized image to acquire current edge coordinates; updating a local coordinate set based on the current edge coordinates acquired; updating the selected area of the to-be-labeled image based on the local coordinate set acquired.

Abstract: A hardware acceleration method and system based on a scale-invariant feature transform algorithm in the field of hardware acceleration design of algorithms is disclosed. The architecture mainly includes two parts: keypoint detection and descriptor generation. Four buffers between these two parts are ping-ponged to increase the system processing speed. In the keypoint detection part, firstly, multi-layer Gaussian pyramid and Gaussian difference pyramid are calculated in parallel; through parallel calculation, the keypoints and gradient magnitudes and orientations are obtained. In the descriptor generation part, a keypoint region division strategy based on the circular keypoint region is provided, and the parallel calculation of a main orientation calculation module and a descriptor generation module are implemented. Finally, the output is obtained through descriptor rearrangement and dimensionality reduction module.

Abstract: A method including generating a plurality of synthetic images of a material, where each synthetic image from among the plurality of synthetic images is associated with a feasibility value greater than a threshold synthetic feasibility value. The method includes determining, for each synthetic image from among the plurality of synthetic images, one or more material properties of the material and one or more process parameters of the material based on the synthetic image and generating a plurality of data points and a pareto surface based on the one or more material properties and the one or more process parameters. The method includes selecting a target data point based on the plurality of data points and a distance between a set of data points from among the plurality of data points and the pareto surface.

Abstract: The present disclosure is related in general to field of machine learning and image processing, that provides a method and system for detecting lane pattern. A lane classification system receives an input image comprising lane markings and co-ordinates of the lane markings, which is divided into two portions, a first image portion and a second image portion based on the lane markings and the co-ordinates of the lane markings. Further, the first image portion and the second image portion are resized into a predefined template size. The lane classification system detects a lane pattern in the two portions by analysing each of the one or more lane markings located in the resized first image portion and the second image portion, independently, based on historical lane data.