Abstract: The present disclosure provides decoding methods, encoding methods, apparatuses and devices. A decoding method includes: obtaining index information of an enhanced temporal motion vector prediction mode from a bit stream of a current block when it is determined that an enhanced temporal motion vector prediction technology is enabled for the current block; determining one or more matching blocks of the current block; determining candidate enhanced temporal motion vector prediction modes based on the one or more matching blocks and one or more new matching blocks obtained by offsetting each of the one or more matching blocks, and establishing a temporal candidate mode list; determining the enhanced temporal motion vector prediction mode based on the index information; determining motion information of each sub-block in the current block based on the enhanced temporal motion vector prediction mode, and performing motion compensation on each sub-block in the current block.

Abstract: Machine learning object-searching methods and apparatuses are disclosed. The method comprises: selecting a state from a set of states of a target object-searching scene as a first state; obtaining a target optimal object-searching strategy whose initial state is the first state for searching for a target object; performing strategy learning by taking the target optimal object-searching strategy as a learning target to obtain an object-searching strategy by which a robot searches for the target object, and adding the obtained object-searching strategy into an object-searching strategy pool; determining whether the obtained object-searching strategy is consistent with the target optimal object-searching strategy; if yes, determining that the strategy learning in which the first state is taken as the initial state of the object-searching strategy is completed; and if not, returning to the step of selecting a state from a set of states of a target object-searching scene.

Abstract: The present disclosure relates to methods, apparatuses and systems for displaying an alarm file and belongs to the field of communications. One of the method includes: through a persistent connection with a service end, receiving a data packet corresponding to the alarm file sent by the service end, wherein the service end releases a storage resource occupied by the alarm file after sending the alarm file; obtaining the alarm file by parsing the data packet corresponding to the alarm file; generating a uniform resource locator (URL) of the alarm file; displaying the alarm file based on the URL of the alarm file. The present disclosure can reduce wastes of the storage resource of the service end.

Abstract: A method for executing a task by a scheduling device, belonging to the technical field of electronics. The method includes: acquiring a target algorithm corresponding to a target task to be executed; acquiring an execution environment condition for a target algorithm, and current execution environment information of various execution devices; in the execution devices, determining a target execution device of which the execution environment information satisfies the execution environment condition; and sending a control message for executing the target task to the target execution device.

Abstract: A fingerprint input method, apparatus, system and an electronic device are provided. The method includes: obtaining a fingerprint as a current fingerprint (S101); comparing the current fingerprint with a fingerprint stored in a fingerprint database (S102); storing, when it is determined that there is no fingerprint identical with the current fingerprint in the fingerprint database, the current fingerprint in the fingerprint database. It can be seen that before storing the obtained current fingerprint into the fingerprint database, it is firstly determined whether there is a fingerprint identical with the current fingerprint in the fingerprint database, and when it is determined that there is no fingerprint identical with the current fingerprint in the fingerprint database, the current fingerprint is stored in the fingerprint database, to avoid the occurrence of duplicate fingerprints in the fingerprint database, avoid the input of the duplicate fingerprints, and improve the user experience.

Abstract: A deep neural network training method and apparatus and a computer device are provided. The deep neural network training method includes: obtaining task attributes of nodes in a current network layer in a tree-like network topology (S101); performing cluster analysis on the nodes in the current network layer based on the task attributes of the nodes in the current network layer and extracting a common part of task attributes of multiple nodes in a same category as a task attribute for a parent node of the multiple nodes (S102); training a network parameter of each parent node based on a task attribute of this parent node (S103); and determining that training of a deep neural network corresponding to the tree-like network topology is completed, after completion of training of all nodes in all network layers (S104). The operation efficiency of deep learning can be improved through this solution.

Abstract: The embodiments of the present disclosure provides a method and an apparatus for training a neural network model, A training sample is obtained, and the neural network model is trained using the training sample. When the neural network model is trained, power exponential domain fixed-point encoding is performed on a first activation inputted into each network layer and a network weight of each network layer, and an encoded first activation and an encoded network weight are power exponential domain fixed-point data, which when used in the operation, can cause a matrix multiplication operation involved to be converted into an addition operation in the power exponential domain by means of the power exponential domain encoding. The hardware resources required for the addition operation are significantly less than that required for the multiplication operation, which therefore can greatly reduce the hardware resource overhead required for running the neural network model.

Abstract: An image processing method and apparatus for target recognition is disclosed. The method includes: inputting N frames of images into a quality evaluation network model; determining, with the quality evaluation network model, a feature vector of each of the N frames of images according to an attention weight of a preset sub-region image and M quality evaluation parameters; determining quality evaluation values of the N frames of images according to the feature vectors of the N frames of images; determining a target image or a target vector for target recognition according to the quality evaluation values of the N frames of images. The feature vector of each frame of image is determined based on the attention weight of the sub-region image and M quality evaluation parameters.

Abstract: The present disclosure provides a decoding method and apparatus, an encoding method and apparatus, and a device.

Abstract: A data processing method and apparatus, and an electronic device, a system and a storage medium are provided. The method is applicable to a main thread in a Web client and includes: obtaining audio and video data of a target video from a server decapsulating the obtained audio and video data to obtain first audio and video data; sending the first audio and video data to a target sub-thread in the Web client, so that the target sub-thread decodes the first audio and video data to obtain second audio and video data and sends the second audio and video data to the main thread; receiving the second audio and video data sent by the target sub-thread; and rendering video data in the second audio and video data by using a rendering module of a browser of the Web client, and taking the rendered video data and audio data in the second audio and video data as to-be-played data for the target video.

Abstract: Disclosed are GPS-based target tracking system, method and dome camera. The system comprises: a first camera and a second camera. The first camera obtains GPS coordinate of a target during the tracking of the target and sends the GPS coordinates to the second camera. The second camera adjusts a shooting angle of the second camera according to the GPS coordinates, and tracks the target after detecting the target in a second monitoring image shot by the second camera. Therefore, in the scheme, if the target moves out of the monitoring range of the first camera, the second camera continues to track the target, namely the tracking range of the target is expanded through linkage of the plurality of cameras.

Abstract: The present application discloses a camera device for collecting vehicle information. In the present application, the camera device uses a dual-lens configuration including a first camera assembly and a second camera assembly, wherein the lens field of view of the first camera assembly intersects with the lens field of view of the second camera assembly, so as to form a layout mode that two areas on both sides of the camera device are photographed in a staggered manner, thereby avoiding the occurrence of a blind area between the lens fields of view of the first camera assembly and the second camera assembly and improving the overall photographing range of the camera device. Moreover, the camera device may introduce at least one auxiliary function such as photosensitive detection, light supplementing, and target detection, so as to improve the quality of image photographing and the accuracy of photographing moment, thereby facilitating improving the accuracy of vehicle information collection.

Abstract: A network monitoring apparatus, and a remote encryption and remote activation method, device and system thereof are provided. The method includes the following steps: receiving an encrypted activation password sent by a client terminal; decrypting the encrypted activation password to obtain an original activation password; determining whether the original activation password meets a predetermined password strength requirement; when the original activation password meets the predetermined password strength requirement, activating the network monitoring apparatus and setting the original activation password as an administrator password; and returning information indicating that the network monitoring apparatus is successfully activated to the client terminal. A network monitoring apparatus, an encryption method of a network monitoring apparatus based on a client terminal, a client terminal, and a remote activation system based on a network monitoring apparatus are also provided.

Abstract: A method and an apparatus for detecting a usage status of a parking space, an electronic device, and a storage medium are disclosed. The method comprises: obtaining a monitored image of a to-be-detected parking space; identifying lane line position information of the to-be-detected parking space from the monitored image; identifying vehicle information from the monitored image, wherein the vehicle information comprises vehicle position information; and determining a usage status of the to-be-detected parking space based on a positional relationship between the lane line position information and the vehicle position information. Determining the usage status of the to-be-detected parking space based on the positional relationship between the lane line position information and the vehicle information is less affected by external environmental factors, which thus may improve the accuracy of a detection result of the usage status of the parking space.

Abstract: The present application provides an encoding and decoding method, apparatus, and device, the method for encoding and decoding includes: determining prediction values of pixel points of current image block; determining gradient values of pixel points of the current image block based on the prediction values of pixel points of the current image block; determining offset vectors of pixel points of the current image block; determining prediction compensation values of pixel points of the current image block based on the gradient values and the offset vectors of pixel points of the current image block; and determining final prediction values of pixel points of the current image block based on the prediction values and the prediction compensation values of pixel points of the current image block. The method can expand the range of application of the prediction compensation adjustment.

Abstract: A method and apparatus for encrypting data includes: acquiring a first data unit included in a media frame to be encrypted, wherein the first data unit includes a boundary portion and a content portion, the boundary portion including boundary information for identifying a starting boundary of the first data unit, and the content portion including content data of the first data unit; encrypting the content data included in the content portion of the first data unit to acquire a second data unit, wherein a boundary portion of the second data unit includes the boundary information, and a content portion of the second data unit includes a data length of the content data and encrypted data acquired by encrypting the content data.

Abstract: A method and apparatus for upgrading software in a front-end device connectable to a back-end device through a coaxial cable. The method includes receiving an instruction for a software upgrade, which is inserted into a horizontal blanking region of a video signal, from the back-end device through the coaxial cable, receiving, in sequence, data packets inserted into the horizontal blanking region, from the back-end device through the coaxial cable, the data packets including data for the software upgrade, and upgrading software in the front-end device based on the received data packets. Existing coaxial cables in a video surveillance system are used to transmit the data for the software upgrade, and during data transmission for the software upgrade the normal monitoring function of the video surveillance system is enabled.

Abstract: Methods and apparatuses for performing object recognition are provided. The method includes: by converting a first feature of a target object extracted using a first extraction model to a feature space of a second extraction model through a first feature conversion model, obtaining a second feature of the target object in the feature space; by matching the second feature of the target object with features of objects in a matching library, obtaining a target feature matched with the second feature of the target object; determining an object to which the target feature belongs as a matching object of the target object. Thus, the present disclosure can save time length of object recognition.

Abstract: The application provides a network camera, a video monitoring system and a method. The network camera comprises: an image sensor, a processor, a memory and a network communication interface; the processor is configured for matching a current image acquired by the image sensor with an image stored in a second storage unit of the memory, so as to obtain a similarity value representing a matching result, and storing an image satisfying a similarity condition in another storage unit of the network camera, reducing the difficulty of image comparison and improving the accuracy of the comparison result.

Abstract: An image capture method and an image capture device are disclosed. The device includes a first chip, zoom lenses, image sensors, zoom motors, and focus motors. The number of the first chip is one; and the number of the zoom lenses are two, comprising a first short zoom lens and a second long zoom lens, wherein the two zoom lenses have different zoom ranges, and there is an overlap range between the zoom ranges of the two zoom lenses. The number of the image sensors is two, and the two image sensors correspond to the two zoom lenses respectively, and the two image sensors are communicatively connected to the first chip respectively. The number of the zoom motors is two, the number of the focus motors is two, and the two zoom motors and the two focus motors are communicatively connected to the first chip respectively.