Abstract: Systems and methods that use at least one neural network to infer content of individual frames in a sequence of images and to further infer changes to content in sequence of images over time to determine whether one or more anomalous events are present in sequence of images is described herein.

Abstract: Systems, methods, and computer program products calculate a historical feature similarity point distribution associated with a first machine learning model, calculate a real-time feature similarity point distribution associated with a second machine learning model, and automatically provide, based on a comparison of the historical feature similarity point distribution to the real-time feature similarity point distribution, an indication of whether the second machine learning model is aligned with the first machine learning model.

Abstract: Provided is a pressure-resistant underwater acoustic coating structure with a mesh structure interlayer, including an outer coating, a mesh structure interlayer, an inner coating, and a metal backing; wherein the outer coating is used for inputting acoustic waves, the mesh structure interlayer and the inner coating are used for consuming acoustic wave energy, and the metal backing is used for enriching a sound absorption mode; and the outer coating, the mesh structure interlayer, the inner coating and the metal backing are attached in sequence.

Abstract: In various examples, a set of object trajectories may be determined based at least in part on sensor data representative of a field of view of a sensor. The set of object trajectories may be applied to a long short-term memory (LSTM) network to train the LSTM network. An expected object trajectory for an object in the field of view of the sensor may be computed by the LSTM network based at least in part an observed object trajectory. By comparing the observed object trajectory to the expected object trajectory, a determination may be made that the observed object trajectory is indicative of an anomaly.

Abstract: Embodiments of the present disclosure provide an interaction method and apparatus, an electronic device, and a storage medium. The method includes: displaying a preset control for target media content on a preset page; and displaying first prompt information in response to a preset trigger operation acting on the preset control, the preset trigger operation is different from a like operation acting on the preset control, the first prompt information is used to prompt that a number of recommendations of media content of a target type is to be increased, and the target type is a media content type of the target media content.

Abstract: Apparatuses, systems, and techniques to train neural networks to perform classification. In at least one embodiment, one or more neural networks are trained to perform classification based, at least in part, on grouping one or more sets of neural network training data according to behaviors of one or more objects within one or more images represented by the training data.

Abstract: The disclosure provides a virtual object processing method, device, and a storage medium. The method includes: First, a virtual scene page corresponding to a target virtual space is displayed, where user identifiers respectively corresponding to a first user and at least one second user included in the target virtual space are presented on the virtual scene page. Then, a first virtual object is determined for the first user based on the virtual scene page. A first virtual resource corresponding to the first virtual object is presented on the virtual scene page in response to receiving a preset trigger operation for the first virtual object, where the first virtual resource includes an animation and a sound effect configured for the first virtual object.

Abstract: In a dynamic spectrum management apparatus, system, method, and computer program product first and second priority level secondary users dynamically use spectrum resources of a primary user. For second priority level secondary users, processing circuitry is configured to, based on blockchain, obtain a spectrum coordination request, acknowledged by a spectrum management apparatus, of another second priority level secondary user serving as a requester for the spectrum resources, wherein the blockchain at least comprises each second priority level secondary user, who is using or requests the use of the spectrum resources, serving as a node, and the spectrum coordination request indicates that the requester makes a request for the second priority level secondary user, who is using the spectrum resources, to perform coordination to provide an interference margin for the requester; and execute a first smart contract on the blockchain, so as to determine content of coordination to be performed.

Abstract: An ECG waveform display method is provided. The method includes: An electronic device starts an ECG waveform detection application, to collect an ECG waveform; the electronic device reads wearing part data between the electronic device and a wrist of a user, where the wearing part data is obtained through IMU wearing part identification; the electronic device determines first wearing part information based on wearing status data, and the determined state is a first determined state or a second determined state; the electronic device obtains second wearing part information between the electronic device and the wrist of the user through ECG wearing part identification; and when the first wearing part information and the second wearing part information are the same and indicate a determined state, the electronic device displays the ECG waveform on a display in a display manner corresponding to the determined state.

Abstract: The field of power systems includes a switching control method for a converter apparatus in a power supply system and the apparatus. This solution may be applied to a power conversion system in scenarios such as a new energy microgrid or an electrochemical energy storage station. In the method, a voltage of the converter apparatus at a control switching moment is stored, the voltage is used as an initial value after switching is performed, and the voltage is transited to a rated voltage in a first time period.

Abstract: An output end of a power converter is connected to a microgrid bus, and the microgrid bus is connected to an external power grid through a grid-tied switch. When detecting that a power grid fails, the power converter is in a first current source control mode. After a duration of the power grid failure reaches a first duration, the power converter switches from the first current source control mode to a first voltage source control mode. The first current source control mode is a current source failure ride-through control mode. The first voltage source control mode is a voltage source failure ride-through control mode. The first duration is less than a second duration, and the second duration is a time interval between a moment at which the power grid fails and a moment at which the grid-tied switch is turned off.

Abstract: Provided are a method and an apparatus of training a radiation image recognition model online and a method and an apparatus of recognizing a radiation image. Preset data is pre-stored in a radiation inspection system, and the preset data includes a first radiation image. The method of training radiation image recognition model online includes: collecting a second radiation image at the radiation inspection site; labeling the second radiation image online to form on-site collected data having a second labeling information, wherein the second labeling information is an incomplete labeling information of the second radiation image, and the second labeling information is used to indicate at least one of objects of interest in the second radiation image; synthesizing the preset data and the on-site collected data online to form a training data set; and training the radiation image recognition model online using the training data set.

Abstract: This application relates to Artificial intelligence and provides a method for training a classifier, one example method including: obtaining a first training sample, where the first training sample includes a corresponding semantic tag; obtaining a plurality of second training samples, where each of the second training samples includes a corresponding semantic tag; determining a target sample from the plurality of second training samples based on semantic similarities between the first training sample and the plurality of second training samples; and training the classifier based on the first training sample, the target sample, and a semantic similarity between the first training sample and the target sample.

Abstract: A device and a method for maintaining the stability of anaerobic ammonium oxidation (anammox) by using a combined zeolite filler to resist water quality fluctuations are provided. The device includes a water inlet tank, an anammox reactor, and a water outlet tank, where the anammox reactor is filled with the combined zeolite filler maintained in a state of suspension, domestic sewage enters the anammox reactor through a water inlet pump, and effluent water enters the water outlet tank through an overflow port. The device combines the physical characteristics of zeolite adsorption of ammonium with the anammox process: when the concentration of ammonium in the influent water is too high, the zeolites will temporarily store the excess ammonium adsorbed; when the concentration of nitrite in the influent water is too high, the zeolites are capable to release the stored ammonium into a liquid phase through balanced adsorption desorption.

Abstract: It is an object of the present invention to provide a method for efficiently producing extracellular vesicles containing biologically active materials such as sonic hedgehog (Shh). The present invention provides, as a solution, a method for producing extracellular vesicles, comprising obtaining extracellular vesicles secreted from cells subjected to a treatment of inactivating Rab GTPase such as Rab18, wherein the cells are preferably cells expressing Shh. In addition, the present invention provides, as a solution, an extracellular vesicle containing Shh and having a particle diameter of 800 nm or more and/or a volume of 109 nm3 or more.

Abstract: Embodiments of this application provide a resource scheduling method and an electronic device, and relate to the field of terminal technologies, to dynamically adjust, on a per-thread basis of a thread included in each application program, a resource allocated by the electronic device to the application program. In the method, a level of each first thread corresponding to a first application program and a scheduling parameter corresponding to each level are set, so that a resource can be allocated to each first thread on a per-thread basis of each first thread included in the first application program. This application is applied to a process in which the electronic device allocates a resource to an application program.

Abstract: Disclosed in the embodiments of the present disclosure are a display control method and apparatus, and a device and a storage medium. The method includes: plays a first video in a first display area in a video stream playing interface of a preset application program. The method further receives a switching instruction. When a preset reminding condition is met, the method switches to play a second video in the first display area. The display area for the second video is changed from the first display area to a second display area during the playing process, and the size of the second display area is smaller than the size of the first display area. In addition, the method displays preset prompt information outside the second display area in the video stream playing interface.

Abstract: A device and method for treating urban domestic sewage based on a two-stage combined process of partial denitrification-anammox belong to the field of biological sewage treatment. The device includes a raw water tank, a sequencing batch biofilm reactor (SBBR), an intermediate water tank, an up-flow anaerobic sludge bed (UASB) and a water outlet tank. A part of urban domestic sewage enters the SBBR and is mixed with residual sewage in the last cycle, a partial denitrification-anammox reaction is carried out under a stirring condition to remove nitrate nitrogen and a part of ammonia nitrogen, followed by a nitrification under an aeration condition to completely convert ammonia nitrogen into nitrate nitrogen, and effluent enters the intermediate water tank; and the other part of the urban domestic sewage is mixed with the effluent of the SBBR and continuously enters the UASB, and nitrite nitrogen, which is generated by nitrate nitrogen reduction, and ammonia nitrogen, are removed by means of anammox.