28 results about "Frame time" patented technology

The application discloses a video instance segmentation method and system based on global feature enhancement, and belongs to the technical field of computer vision and deep learning, which comprises the following steps: inputting continuous video frames into a double-channel decoupling feature extraction module to obtain appearance and position features; inputting the two types of features into a tracker and a global feature memory updating module to dynamically update and obtain long-time sequence global context features; completing instance detection matching by the tracker, judging target state to obtain disappeared and newly appeared target queries; combining long-time sequence features to model current frame time sequence to generate enhanced features; finally, the tracker completes segmentation and tracking based on the enhanced features and the target queries, and outputs a segmentation mask sequence, an instance ID, a category ID and a tracking state. Through double-channel decoupling, double-memory block updating, multi-cost matching and multi-weight feature fusion, the application strengthens appearance feature representation and long-time sequence information utilization, and significantly improves segmentation accuracy and tracking robustness in complex scenes.

The application relates to the technical fields of multimedia image processing, GPU computing power scheduling and soft power-on resource optimization, and discloses a resource occupation prediction method in a video special effect processing process, which is executed by a terminal built-in processor and comprises the following steps: collecting a video special effect frame time interval, a GPU rendering pipeline thread occupation ratio and a video memory fragment distribution ratio; performing dimensionless normalization processing on the video special effect frame time interval to obtain a special effect frame time normalization factor; counting a special effect superposition level; and calculating a special effect time coupling factor, a pipeline blocking loss factor and a total resource occupation prediction value and outputting the total resource occupation prediction value. The application solves the problem of large prediction deviation in the prior art, realizes accurate prediction through a core innovation point, guarantees reasonable allocation of video special effect rendering resources, and improves rendering stability.

Example embodiments relate to taking images at certain predetermined angles in order to have consistent exposure throughout the images. An example embodiment includes a method. The method includes determining, using a lidar device, light intensity information of a surrounding environment of the lidar device. The light intensity information includes a plurality of angles within a threshold range of light exposure. The method also includes determining rotation times associated with each of the angles within the threshold range of light exposure. Further, the method includes based on the rotation times associated with each of the angles within the threshold range of light exposure, determining a plurality of target image times. In addition, the method includes capturing, by a camera system, a plurality of images at the plurality of target image times.

The present application relates to the technical field of intelligent control and industrial visual detection of beneficiation process, and discloses an industrial-grade ore belt boundary identification method based on multi-frame time sequence fusion. A multi-branch hollow boundary enhancement unit is constructed based on deep convolution, different convolution kernel sizes, and a Sobel operator. A feature extraction and fusion module is constructed based on unbiased convolution, time sequence cross-channel difference, the multi-branch hollow boundary enhancement unit, and multi-head attention. A feature enhancement module is constructed based on deep separable convolution and the multi-branch hollow boundary enhancement unit. A prediction module is constructed based on the multi-branch hollow boundary enhancement unit. An identification model is constructed based on the feature extraction and fusion module, the feature enhancement module, and the prediction module. Continuous multi-frame pictures in an industrial field are acquired, and after preprocessing, the multi-frame pictures are input into the identification model to obtain ore belt boundary identification results. Based on the ore belt boundary identification results, sorting is performed, and the problem that existing table sorting and identification methods cannot effectively identify ore belt boundaries is solved.

This invention discloses a rendering optimization method, program product, and storage medium based on frame interpolation and super-resolution. The method includes: acquiring an optimized super-resolution rendering frame at time t as a first rendering frame, and acquiring a low-resolution rendering frame at time t+1 in the frame display sequence as a second rendering frame; reprojecting the first rendering frame and the upsampled second rendering frame to an intermediate frame time and correcting them together to obtain corrected data for a high-resolution intermediate frame; reprojecting the features of the first and second rendering frames to the intermediate frame time and combining them with the features of the corrected data for the high-resolution intermediate frame to reconstruct a super-resolution intermediate rendering frame; reprojecting the first rendering frame to the time of the second rendering frame and correcting it together with the upsampled second rendering frame to obtain corrected data for the second rendering frame; extracting features from the corrected data for the second rendering frame; and reconstructing the super-resolution rendering frame at time t+1 based on the features of the corrected data for the second rendering frame and the features of the second rendering frame. This invention has a short running time and good rendering effect.

The present invention relates to a method and apparatus for transmitting messages. A method for transmitting a message onto a time sensitive network, wherein the method comprises: receiving a message on a plurality of different channels; determining a frame time required to transmit a first received message from a first channel onto the network. If the frame time is greater than a predetermined maximum time period, maxInterferenceTime, permitted to transmit a single message within the network, then determining a minimum time, waitTime, until a second message on a second channel can become available for transmission. If the waitTime is greater than or equal to the difference between the frame time and the maxInterferenceTime, then transmitting the first message.

The application relates to a multi-camera asynchronous acquisition VIO implementation method, comprising the following steps: acquiring the timestamps of multiple images corresponding to a current frame; wherein the multiple images are images captured by multiple cameras within a same frame time window; determining the timestamp of a reference window corresponding to the current frame according to the timestamps of the multiple images; calculating the pose compensation amount between the timestamp of each image and the reference window based on inertial measurement data and a state vector; respectively constructing inertial measurement residuals and visual measurement residuals corresponding to each image based on the pose compensation amount; and iteratively optimizing and solving the inertial measurement residuals and the visual measurement residuals based on a nonlinear optimization method. The scheme of the application models the inconsistent difference amount of the timestamps of the pictures taken by different cameras in the VIO system, so that when the different cameras in the VIO system do not take pictures at the same time, higher precision and robustness can be achieved; the scheme can make the VIO system have higher detection precision and higher efficiency for a rotating target.

This invention discloses a 3D visualization and intelligent control platform, relating to the field of 3D visualization technology. It constructs a continuous temporal buffer zone for the 3D flow field, rearranging the time indices of adjacent frames to form a sliding interval before transient switching. Based on this interval, a flow direction matching window is embedded, and streamlines are corrected using inter-frame velocity vector differences, introducing a directional channel. Accordingly, a local anti-phase shielding layer is set to suppress anti-phase rendering. An energy consistency path is established between the rendering layer and the control logic layer to achieve visual and logical flow direction unification. Control commands are issued with a delay according to the actual flow direction rhythm through a rhythmic ventilation mapping chain. This invention, by constructing a continuous temporal buffer zone, a flow direction matching window, and an energy consistency path, achieves time and direction synchronization of the 3D flow field, eliminates phase drift and anti-phase rendering, and dynamically unifies the rendering layer and the control logic layer, improving the visual stability, control accuracy, and operational safety of the ventilation system under complex operating conditions.

The application discloses a data transmission method, device, equipment, medium and program product. The method comprises the following steps: acquiring an environment state parameter of to-be-transmitted information, wherein the environment state parameter at least comprises a state parameter of a receiving end, and the state parameter of the receiving end comprises a buffer capacity of the receiving end, a freeze frame time length of the receiving end and a signal reconstruction quality of the receiving end; inputting the environment state parameter into a first forward error correction (FEC) unequal error protection strategy model to obtain an FEC protection rate corresponding to the environment state parameter; performing FEC packetization on the to-be-transmitted information according to the FEC protection rate to obtain target transmission information; and performing transmission on the target transmission information based on a transmission link. Stable high-quality data transmission is realized.

The application discloses a framing time and bandwidth allocation joint optimization method of an emergency communication network, under a time-varying channel state, a state space, an action space and a reward function of the emergency communication network are constructed; a double-depth Q network is constructed, including an online network and a target network; experience samples generated by interaction of an agent and an emergency communication network environment are stored in an experience replay pool; for each experience sample, based on an experience enhancement extraction mechanism, a sampling priority, a sampling probability and an importance sampling weight are calculated; a preset number of experience samples are randomly extracted, the agent is trained, and the agent outputs an optimal action according to a real-time state of the emergency communication network, so that link resource allocation of the emergency communication network is completed. Through real-time interaction of the agent and the environment, the application autonomously learns an optimal combination strategy of the framing time and the bandwidth allocation, so that under a time-varying channel condition, the link bandwidth utilization is significantly improved and the data transmission delay is cooperatively reduced.

This invention discloses an azimuth estimation method based on frame weighting and frequency sliding generalized cross-correlation. The method acquires multi-channel microphone received signals and obtains time-frequency domain observation signals through short-time Fourier transform. The time-frequency domain observation signals are decomposed and de-reverberated to obtain enhanced signals. Based on the enhanced signals, a time difference generalized cross-correlation vector is calculated, and all vectors are stacked to construct an FS-GCC matrix. After singular value decomposition and rank-1 approximation, the single-frame time difference estimation results are extracted. Frame weighting coefficients are calculated by combining the time-frequency domain observation signals and the enhanced signals, and weighted fusion is performed to obtain the final time difference estimate. The azimuth and elevation angles of the far-field sound source are obtained using the least squares method. This invention, by combining frequency sliding and frame weighting processing, effectively improves the accuracy of sound source arrival estimation in reverberant and noisy environments, and has good engineering application value.

The application provides a harmful video detection method and device based on cross-frame timing semantic understanding. Based on the feature that harmful semantics in a video is often jointly constituted by entity attributes, entity relationship and timing evolution process, a hierarchical analysis and joint reasoning strategy for complex semantic structure is proposed. A hierarchical modeling and joint decision technology is adopted to perform harmful semantic analysis from three aspects of entity attribute and emotion modeling, entity timing relationship modeling and global semantic graph reasoning, and finally determine through semantic fusion, so as to improve the recognition ability of implicit harmful content and rule violation semantics in complex context.

This invention provides a few-shot action recognition network, method, device, and storage medium based on meta-learning and time alignment. It includes a meta-training unit and a meta-testing unit. The meta-training unit trains an embedding model with good generalization performance to extract temporal and spatial information from video frames. After training, all network headers, including the classifier, are removed. η The embedded model / feature extractor, whose parameters are frozen or fine-tuned as a meta-test unit, is then evaluated by a linear classifier. This invention applies the novel meta-learning algorithm NIL to few-sample action recognition, achieving accurate and efficient classification of human actions with limited sample data.

This application provides a method, apparatus, receiver, and storage medium for synchronizing the incoming reference time standard of an RDSS receiver. The method includes: determining the timing interrupt time difference by acquiring tracking channel observation data and correcting the alignment of the timing interrupt trigger time with the outgoing reference time standard; simultaneously acquiring the outgoing frame number from the satellite broadcast and maintaining the outgoing frame time count and the incoming reference time standard during each timing interrupt; ensuring that the incoming response time can be accurately determined and controlling the transmission module to send the incoming signal on time; and solving the technical problems of inaccurate incoming response transmission timing caused by difficulty in aligning the timing interrupt with the outgoing reference time standard and easy drift in intra-frame time maintenance in existing counting methods.

A method for performing media playback on a media playback device, including: receiving a data stream to buffer the data stream in a buffer unit or to record the data stream in a storage unit; performing a parsing operation on the buffered data stream or the recorded data stream to obtain frame time and data offset corresponding to one or more intra-coded pictures of at least one audiovisual (AV) content included in the data stream, thereby to create a frame index table; in response to a playback operation, referring to the frame index table to determine a data offset associated with the playback operation; retrieving one or more data units from the buffer unit or the storage unit according to the data offset; and decoding the one or more data units for media playback.

A control method and system, a tracking method and system, a device, and a storage medium are provided. The control method includes: acquiring, using a detector (200), an image of a movable device (100) at a current frame time, the movable device (100) being equipped with a plurality of light output units (110) for outputting signal light (S1); predicting, based on the image, a motion state of the movable device (100) at a next frame time (S2); and obtaining, based on the motion state of the movable device (100) at the next frame time, configuration information of a light output state of each of the light output units (110) at the next frame time, to control the light output unit (110) (S3). The present invention is conductive to adaptively adjusting the light output states of all the light output units (110) based on a real-time state of the movable device (100), which facilitates the light output units (110) to achieve accurate and low-power light output states to ensure the use accuracy of the movable device (100) and reduce the power consumption of the movable device (100).

The application provides a frequency modulation method, a frequency modulation device and electronic equipment, and relates to the technical field of processors. The method comprises the following steps: for a target service, determining a plurality of target threads running on a heterogeneous multi-core processor, and obtaining historical running data of the plurality of target threads; determining a thread reserved running time of a next frame corresponding to each target thread according to a frame time difference of the target service; wherein the frame time difference is a difference between an actual frame length of the frame and a target frame length corresponding to the target service; based on the historical running data, predicting a predicted running time and a predicted performance index corresponding to the target thread when running at different frequencies of a candidate kernel type combination; selecting a target kernel type combination and a corresponding target frequency combination that meet the requirements of the predicted running time and the predicted performance index, so as to run the target thread in the next frame. Based on this, frame-granularity precise frequency modulation control and thread kernel distribution adjustment are realized.

This application provides a game server scheduling method, apparatus, and electronic device. The method includes: responding to a microservice request received during the main game loop's processing of the current logical frame; predicting the remaining frame time of the current logical frame and the expected response latency of the microservice request; the remaining frame time refers to the available processing time not yet consumed within the processing cycle of the current logical frame; the expected response latency refers to the time elapsed from the initiation of the microservice request to its completion; selecting a target scheduling strategy from multiple preset scheduling strategies based on the relative relationship between the remaining frame time of the current logical frame and the expected response latency of the microservice request; and allocating a target thread to process the microservice request according to the target scheduling strategy, so that the target thread can process the microservice request. This application solves the technical problem of game "frame skipping" or "stuttering" caused by asynchronous requests from non-core microservices blocking the main game loop.

The embodiment of the invention discloses a multi-channel event video synchronous playback method and system, and relates to the technical field of video monitoring. The method comprises the following steps: creating a global synchronization control instance and a plurality of decoding playing instances, wherein each instance corresponds to a video channel and is registered to the synchronization control instance; and establishing an independent video frame buffer queue for each channel. And starting the decoding playing instance to enter a decoding operation state, and sending a rendering request containing a timestamp of a frame to the synchronous control instance before the video frame obtained by decoding is rendered. The synchronization control instance executes synchronization control logic and compares the timestamp with a globally maintained current playing timestamp. And through centralized arbitration, the corresponding decoding playing instance is approved to render the frame only when the difference value of the two is within the preset fault-tolerant range. According to the method and the device, frame-by-frame synchronous control on multi-channel video playback is realized, each channel is forced to advance according to a unified timeline, and strict synchronization and experience continuity of cross-channel playback pictures are ensured.

The method comprises the following steps: after detecting that a target cloud game is started, obtaining a target frame timestamp and determining a frame rate fluctuation coefficient; determining an input data interpolation frame mode of the target cloud game according to the target frame timestamp and the frame rate fluctuation coefficient; determining interpolation data frame information of the target cloud game according to the input data interpolation frame mode and the buffered information in a created data buffer queue; and updating game picture information of the target cloud game according to the interpolation data frame information. By using the method, the interpolation data frame as the input frame can be determined according to the received timestamp of the information input frame and the frame rate fluctuation coefficient in the cloud game running, and the game picture information is formed through the interpolation data frame. The processing mode guarantees the continuity of the game input information, better reduces the input delay, and thus improves the operation accuracy and smoothness of the player on the cloud game.

The present invention relates to methods and apparatus for transmitting messages. A method for transmitting messages onto a time sensitive network. The method comprises receiving messages on a plurality of different channels; determining a frame time required to transmit a first received message from a first channel onto the network. If the frame time is greater than a predetermined maximum time period, maxInterferenceTime, permitted to transmit a single message within the network, then: splitting the first message into two or more sub-messages; determining a time required to transmit each sub-message onto the network. If the time required to transmit each respective sub-message is less than or equal to the maxInterferenceTime, then each sub-message is sequentially transmitted onto the network in turn, enabling a second message from a second channel to be transmitted between sub-messages if required.

This electronic device comprises: a communication interface; a display; at least one processor; and a memory comprising at least one storage medium for storing instructions. The instructions, when executed individually or collectively by the at least one processor, cause the electronic device to: obtain, via the communication interface, a series of encoded source frames, a respective source frame duration for each source frame, and first reference point timing information for a first reference point defined on a first source frame among the series of source frames; generate playback frames for playback via the display by decoding the source frames; determine a playback delay value on the basis of the first reference point timing information and the respective source frame durations of source frames corresponding to playback frames that have been played back prior to the acquisition of the first reference point timing information; and perform delay control on the basis of the playback delay value.

This invention provides a method, system, and electronic device for implementing 2D game physics gravity trajectory. The method includes the following steps: configuring physical environment attributes for the game scene; configuring initial physical attributes for physical objects in the game; calculating the acceleration of the physical object in the current frame based on the physical environment attributes, initial physical attributes, and the current velocity of the physical object; updating the position and velocity of the physical object in the update callback of each frame of the game based on the acceleration and frame time interval, and storing the updated position coordinates in the trajectory list; repeating the above steps until trajectory information composed of position coordinates of multiple consecutive frames is obtained; binding the current position coordinates of the physical object to the corresponding display element and displaying it on the device screen. This method can accurately simulate the motion trajectory under the combined action of gravity, drag, and wind in a real physical environment, improving the realism and immersion of the physics simulation.

This invention relates to the field of display module technology and discloses a wireless multi-card synchronous control method for display modules and a display screen. The method includes: acquiring a first frame timestamp through a master control card and encapsulating the first frame timestamp into a frame synchronization broadcast packet for wireless broadcast within the time slot corresponding to the dedicated transmission time slot offset; acquiring a second frame timestamp through multiple slave control cards and performing step correction on the first local frame period until the frame phase error converges to within the phase-locked steady-state threshold to obtain the second local frame period; each slave control card drives its subordinate display module with the second local frame period, and the master control card reports the frame phase-locked state status response packet according to the time slot corresponding to the dedicated transmission time slot offset of each slave control card. This invention ensures that the entire network continuously maintains phase-locked steady state under dynamic disturbances, thereby realizing reliable wireless synchronous refresh control of large-scale multi-card display modules without external synchronization cables.