Abstract: A method for detecting abnormal scene may include obtaining data relating to a video scene, identify at least two motion objects in the video scene based on the data and determining a first motion feature relating to the at least two motion objects based on the data. The method may also include determining a second motion feature relating to at least one portion of each of the at least two motion objects based on the data. The method may further include determining whether the at least two motion objects are involved in a fight based on the first motion feature and the second motion feature.

Abstract: This invention proposes a pedestrian re-identification method based on virtual samples, comprising following steps: s1) obtaining virtual persons generated by game engine, and generating the virtual samples with person labels by fusing a background of a target dataset and a pose of real persons through a multi-factor variational generation network; s2) rendering the generated virtual samples according to lighting conditions; s3) sampling the rendered virtual samples according to person attributes of target dataset; s4) constructing a training dataset according to virtual samples obtained by sampling to train a pedestrian re-identification model, and verifying identification effect of the trained model.

Abstract: A system and a method for an electronic device for counting objects of similar type in an area includes a register for storing a number of count of the objects in the area; a depth sensor for measuring depth distances between a plurality of points of an object and the depth sensor; a thermal sensor for measuring temperatures of the points of the object; wherein the distances and the temperatures are used to derive spatial thermal information of the object for determining an increase in the number stored in the register.

Abstract: A control device in a premises security system for a premises is provided. The control device is configured to receive video surveillance data associated with an area of the premises, identify, using at least one machine learning model, a triggering event based at least in part on the video surveillance data, identify, using the at least one machine learning model, and based at least in part on the video surveillance data, a person associated with the triggering event, identify, using the at least one machine learning model, at least one characteristic of the person associated with the triggering event, generate an audio message comprising content based at least in part on the at least one characteristic of the person associated with the triggering event, and cause playback of the audio message in the area of the premises.

Abstract: Described is a multiple-camera system and process for re-identifying a user located in a materials handling facility based on user patterns and/or descriptors representative of the user. In one implementation, a user pattern and/or a plurality of descriptors representative of a user are maintained as a position of a user is tracked through a materials handling facility. If the tracking of the user is lost, the last known position is stored with the user pattern and/or descriptors. If a new object is detected and confirmed to be a user, a user pattern and/or descriptors of the new object are compared with the stored user pattern and/or descriptors to determine if the new object is the user.

Abstract: Disclosed are techniques for augmenting video datasets for training machine learning algorithms with additional video datasets that are cropped copies of the video datasets. Frames of a received video dataset are divided into a plurality of subframes. For each subframe, a count is tallied corresponding to the cumulative number of pixels changed across the frames of the received video. Counts are compared to determine which subframe includes the most changed pixels across the frames of the video dataset, which is selected as a cropping candidate. The cropping candidate is used to generate copies of the video dataset that are cropped to include at least the cropping candidate and exclude at least some of the remaining portions of each frame of the video dataset that are outside of the cropping candidate. In some embodiments, boundaries of cropping candidates are transformed to generate a plurality of cropped variations of the video dataset.

Abstract: Present embodiments pertain to systems, apparatuses, and methods for analyzing and reporting rotational or reciprocating movements and speed of rotation or reciprocation in mechanical structures, machines, and machine components, through the use of an acquired video recording representing a plurality of cycles of motion, by measuring intensity values of a subset of pixels contained in a region of interest within the video recording in a plurality of frames of the video recording, thereby calculating a speed of rotation or reciprocation of the structure, machine or machine component.

Abstract: A method of operating a server providing a sports video-based platform service includes tracking a ball by using a pre-processing of leaving dynamic pixels in the sports video of a ball game, determining an unidentified player associated with a score-related scene of the sports video by using a ball tracking result, identifying the unidentified player by tracking the unidentified player up to an adjacent frame where the unidentified player is identifiable, and generating a time section of the sports video and identification information for the unidentified player, corresponding to the score-related scene.

Abstract: An image processing apparatus obtains first image data as correct image data, and second image data generated by deteriorating image quality of the first image data as input image data, and generates learning data that links the correct image data with the input image data, and learns a neural network using the generated learning data, so that the image processing apparatus generates a learning model used to convert input image data.

Abstract: An object display method, an object display apparatus, an electronic device, and a computer readable storage medium are provided. The object display method comprises: receiving video images; recognizing a target object in the video image; displaying, in response to a first target object being recognized, a first object having a first shape at a first location in the video images; displaying, in response to a second target object being recognized, a second object having a second shape at a second location in the video images; and combining, when a distance between the first target object and the second target object is less than a first threshold, the first object and the second object so that the first object and the second object form a third shape.

Abstract: A method performed by an electronic device is described. The method includes obtaining a first image from a first camera, the first camera having a first focal length and a first field of view. The method also includes obtaining a second image from a second camera, the second camera having a second focal length and a second field of view disposed within the first field of view. The method further includes aligning at least a portion of the first image and at least a portion of the second image to produce aligned images. The method additionally includes fusing the aligned images based on a diffusion kernel to produce a fused image. The diffusion kernel indicates a threshold level over a gray level range. The method also includes outputting the fused image. The method may be performed for each of a plurality of frames of a video feed.

Abstract: A certificate image extraction method, including: step S101, obtaining an original image containing a certificate image, wherein the original image is obtained by a camera device by means of photographing; step S102, performing white balance processing on the original image to obtain a balance image according to component values of pixel points in the original image in red, green and blue color components; step S103, determining a position of the certificate image in the balance image according to a pre-trained certificate feature model; wherein the certificate feature model is obtained by training based on historical certificate images, a certificate image model and a preset initial weight value; and step S104, extracting the certificate image from the balance image according to the position of the certificate image. By performing the certificate image extraction method, the accuracy of extracting the certificate image from the original image is improved.

Abstract: Methods, systems, and apparatus for ground plane filtering of video events are disclosed. A method includes obtaining a first set of images of a scene from a camera; determining a ground plane from the first set of images of the scene; obtaining a second set of images of the scene after the first set of images of the scene is obtained; determining that movement shown by a group of pixels in the second set of images of the scene satisfies motion criteria; determining that the ground plane corresponds with at least a portion of the group of pixels; and in response to determining that movement shown by the group of pixels in the second set of images of the scene satisfies motion criteria, and that the ground plane corresponds with at least a portion of the group of pixels, classifying the group of pixels as showing ground plane based motion.

Abstract: Systems and methods for efficiently training a machine learning model are presented. More particularly, using information regarding the relevant neighborhoods of target nodes within a body of training data, the training data can be organized such that the initial state of the training data is relatively easy for a machine learning model to differentiate. Once trained on the initial training data, the training data is then updated such that differentiating between a matching and a non-matching node is more difficult. Indeed, by iteratively updating the difficulty of the training data and then training the machine learning model on the updated training data, the speed that the machine learning model reaches a desired level of accuracy is significantly improved, resulting in reduced time and effort in training the machine learning model.

Abstract: A computer-implemented method and system provide the ability to determine and provide a safety risk analysis for construction. Construction related data is obtained and includes textual data and a visual artifact for the construction project. A construction safety context is identified based on the construction related data. Based on the construction safety context, a safety participant risk score that assigns a numerical safety risk participant value to any entity involved in the construction project is determined. Based on the safety risk participant score, a safety project score that assigns a risk level on a per-project basis is determined. The safety risk analysis is presented based on the safety participant risk score and safety project score, via a graphical user interface.

Abstract: A method for identifying a pore color of at least one skin pore. The method includes obtaining at least one image of a subject, extracting a pore image from the obtained image, in which the extracted pore image identifies a boundary of the at least one skin pore, and superimposing the obtained image and the extracted pore image to identify the pore color within the identified boundary of the at least one skin pore.

Abstract: An image processor and a method therein to provide a target image for evaluation with an object detector. The method comprises: obtaining a source image captured by a camera and depicting an object, and applying an inverse pixel transform to each target pixel of a target image to determine one or more source pixels located at a position in the source image corresponding to a position of each target pixel in the target image. The method further comprises assigning, to each target pixel, a target pixel value determined based on one or more source pixel values of the determined one or more source pixels located at the corresponding position, thereby is a size c in the target image of the depicted object of a specific object type normalized in at least one size dimension. Thereafter, the target image is fed to an object detector for evaluation.

Abstract: Disclosed is a system for assisting in guiding and performing a procedure on a subject. The subject may be any appropriate subject such as inanimate object and/or an animate object. The guide and system may include various manipulable or movable members, such as robotic systems, and may be registered to selected coordinate systems.

Abstract: An object tracking system includes a first sensor, a second sensor, and a tracking system. The first sensor is configured to capture a first frame of a global plane for at least a first portion of a space. The second sensor is configured to capture a second frame of at least a second portion of the space. The tracking system is configured to determine the object is within an overlap region with the second sensor based on a first pixel location. The tracking system is further configured to determine a first coordinate in the global plane for the object, to determine a second pixel location in the second frame for the object based on the first coordinate, and to store the second pixel location with an object identifier a tracking list associated with the second sensor.

Abstract: An image segmentation method, apparatus, and a storage medium are provided. The method includes: selecting a current image frame in a video according to a time sequence of the video; determining a reference image frame before the current image frame in the time sequence; obtaining first location information of a target object key point in the reference image frame; performing an affine transformation on the current image frame with reference to an affine transformation relationship between the first location information and a target object key point template to obtain a target object diagram; performing a key point detection on the target object diagram to obtain second location information of the target object key point; segmenting a target object from the target object diagram to obtain segmentation information; and obtaining the target object from the current image frame according to the segmentation information and the second location information.