Peking University

Zte Corporation

Zhenyu Wang

Ronggang Wang

Wen Gao

Wenmin Wang

Shengfu Dong

Ge Li

Ying Li

Dajiang Wang

Hui Zhao

Yueming Wang

Yang Zhao

Tiejun Huang

Xinxin Zhang

Kui Fan

Mengdi Fan

Youdao Ye

Siwei Ma

Yihao Zhang

Jun Dong

Xiubao Jiang

Mingmin Zhen

Peilei Dong

Ruonan Zhang

Lei Chen

Lin Li

Sheng Wang

Hongbin Cao

Qilei Wang

Xi Xie

Zhihong Kang

Zhihao Li

Chengzhou Tang

Hao Lv

Yuqiang Lai

Xikun Yang

Guisen Xu

Chuanxin Tang

Jianbo Jiao

Chengzan Zhang

Lu Ran

Weiqing Li

Haoxian Zhang

Hongmeng Song

Liang Han

Shen Zhang

The present disclosure provides a method for optimizing OTN resources, including: determining and creating, according to an action policy, a service to be created in a current service creation state, calculating a timely reward in the current service creation state, entering a next service creation state until an episode is ended, and calculating and updating, according to the timely reward in each service creation state, an optimized objective policy parameter in each service creation state; iterating a preset number of episodes to calculate and update the optimized objective policy parameter in each service creation state; determining, according to the optimized objective policy parameter in each service creation state in the preset number of episodes, a resultant optimized objective policy parameter in each service creation state; and updating the action policy according to the resultant optimized objective policy parameter in each service creation state.

Method for optimizing OTN resources, computer device and storage medium

The present disclosure provides a method for optimizing OTN network resources, including: determining and creating a service to be created in a current service creating state according to an action policy, calculating a timely reward, entering a next service creating state, until an Episode is finished, calculating a comprehensive optimization parameter according to the timely reward, and calculating and updating a quantization index weight vector according to the comprehensive optimization parameter, where the action policy is a probability function related to the quantization index weight vector, and the quantization index weight vector corresponds to a plurality of quantization indexes; iterating a preset number of Episodes to obtain an optimized quantization index weight vector; and updating the action policy according to the optimized quantization index weight vector. The present disclosure further provides an apparatus for optimizing OTN network resources, a computer device, and a computer-readable storage medium.

Method and apparatus for optimizing OTN network resources, computer device and storage medium

A network management and control method and system, a network system, and a storage medium are disclosed. The network management and control method may include: acquiring Digital Twin (DT) model data corresponding to physical objects in a physical network (S); generating, according to the DT model data, DT cases at different levels, wherein the levels of the DT cases correspond to levels of the physical objects, and the DT cases at different levels have a functionally collaborative relationship (S); obtaining a target analysis result according to the collaborative relationship and all the DT cases (S); and generating network configuration information according to the target analysis result, and delivering the network configuration information to the physical network so that the physical network implements network management and control according to the network configuration information (S).

Network management and control method and system, network system, and storage medium

An information processing method, a method for generating and training a module, an electronic device, and a medium are disclosed. The method for information processing may include acquiring a transmission parameter of service information in a network, where the transmission parameter is indicative of transmission performance of the service information in the network; and predicting transmission performance of the transmission parameter by means of a pre-trained network model, where the network model comprises a Graph Convolutional Network (GCN) model that is acquired according to transmission path information of the service information in the network.

Information processing method, method for generating and training module, electronic device, and medium

Embodiments of the present disclosure provide a method and device for selecting a context model of a quantized coefficient end flag. The method comprises: obtaining a scanning position POS of a non-zero coefficient corresponding to current quantized coefficient end flag in a specific scanning order; wherein the scanning position POS is a subscript of the non-zero coefficient in the scanning order; configuring a first context model array, and using a fixed value as the base to calculate the logarithmic value of the value of the scanning position POS plus 1, and according to the logarithmic value, selecting a first context model from the first context model array; and using the first context model to encode or decode a binary symbol of the current quantized coefficient end flag. According to the technical solution of the present application, it is able to improve encoding and decoding efficiency for a quantized coefficient end flag, thereby further improve the efficiency of video encoding and decoding.

Method and device for selecting context model of quantization coefficient end flag bit

An embodiment of the present description provides a decoding method and device for a quantization block as well as an electronic device. The method comprises: determining the scanning order of quantization coefficients in the quantization block, initializing the scanning position of the quantization coefficients, and decoding a run-length; adding a subscript value of the current scanning position and the value of the run-length to obtain the scanning position of nonzero coefficients, decoding the values of the nonzero coefficients; judging whether the subscript value of the nonzero coefficients is the maximum subscript value, if yes, ending the decoding of the quantization block; if not, judging whether there are undecoded nonzero coefficients, and if there are no undecoded nonzero coefficients, ending the decoding of the quantization block; if there are undecoded nonzero coefficients, adding 1 to the subscript value of the decoded nonzero coefficients to obtain an updated current scanning position, taking the current scanning position as the initial scanning position to decode the next run-length until the decoding of the quantization block is completed. The technical solution of the present application can improve the decoding efficiency of quantization block.

Decoding method and device for quantization block, video decoding method, and electronic device

A method, system, device and computer-readable storage medium for inverse quantization. The method comprises: determining an initial weighted inverse quantization matrix, wherein, the initial weighted inverse quantization matrix is the same as the quantized block in size; setting some matrix elements in the initial weighted inverse quantization matrix to zero to obtain a weighted inverse quantization matrix, wherein, determining the matrix elements that need to be zeroed according to the size of the quantized block; weighted inverse quantizing the quantized coefficients in the quantized block to generate corresponding inverse transform coefficients, wherein, the value of the matrix element corresponding to the position of the quantized coefficient in the weighted inverse quantization matrix is used as a weight coefficient of the weighted inverse quantization.

Method, system, device, and computer-readable storage medium for inverse quantization

The present application discloses a cross-media retrieval method based on deep semantic space, which includes a feature generation stage and a semantic space learning stage. In the feature generation stage, a CNN visual feature vector and an LSTM language description vector of an image are generated by simulating a perception process of a person for the image; and topic information about a text is explored by using an LDA topic model, thus extracting an LDA text topic vector. In the semantic space learning phase, a training set image is trained to obtain a four-layer Multi-Sensory Fusion Deep Neural Network, and a training set text is trained to obtain a three-layer text semantic network, respectively. Finally, a test image and a text are respectively mapped to an isomorphic semantic space by using two networks, so as to realize cross-media retrieval. The disclosed method can significantly improve the performance of cross-media retrieval.

Cross-media retrieval method based on deep semantic space

A video action detection method based on a convolutional neural network (CNN) is disclosed in the field of computer vision recognition technologies. A temporal-spatial pyramid pooling layer is added to a network structure, which eliminates limitations on input by a network, speeds up training and detection, and improves performance of video action classification and time location. The disclosed convolutional neural network includes a convolutional layer, a common pooling layer, a temporal-spatial pyramid pooling layer and a full connection layer. The outputs of the convolutional neural network include a category classification output layer and a time localization calculation result output layer. The disclosed method does not require down-sampling to obtain video clips of different durations, but instead utilizes direct input of the whole video at once, improving efficiency. Moreover, the network is trained by using video clips of the same frequency without increasing differences within a category, thus reducing the learning burden of the network, achieving faster model convergence and better detection.

Video action detection method based on convolutional neural network

The embodiment of the present specification discloses an image synthesis method, apparatus and device for free-viewpoint. The method includes: correcting a first depth map based on a first color map and the first depth map inputted, wherein the first color map includes a left image and a right image, and the first depth map includes a left image and a right image; forward-projecting the second depth map to a virtual viewpoint position based on the pose of a virtual viewpoint to obtain a third depth map, where the third depth map is a projection map located at the virtual viewpoint position; back-projecting the left and right color maps closest to the virtual viewpoint position in the first color map to the virtual viewpoint position based on the third depth map to obtain a second color map, where the second color map is a projection view of the first color map at the virtual viewpoint position; correcting the second color map by using an optical flow algorithm to obtain a third color map, where the optical flow algorithm implements aligning the left and right images of the second color map; performing weighting calculation and blendering on the third color map to obtain a composite image.

Image synthesis method, apparatus and device for free-viewpoint

The present specification discloses a method, apparatus, and device for video frame interpolation. The method of embodiment of the present specification comprises: acquiring a video frame training sample, wherein the video frame training sample includes an even number of consecutive video frames and a first key frame, and the first key frame is an intermediate frame of the even number of consecutive video frames; constructing a pyramid deep learning model, wherein each level of the pyramid deep learning model being used to generate intermediate frames of different resolutions has a plurality of convolutional neural network layers; inputting the even number of consecutive video frames to the pyramid deep learning model to generate a second key frame; modifying the pyramid deep learning model according to the second key frame and the first key frame to generate a modified pyramid deep learning model; inputting a plurality of video frames to be processed into the modified pyramid deep learning model to generate an intermediate frame of the plurality of video frames. The invention fully exploits the spatio-temporal domain information between multi-frame video frames, and adopts a pyramid refinement strategy to effectively estimate the motion information and the occlusion region, thereby greatly improving the quality of the intermediate frame.

Method, apparatus, and device for video frame interpolation

A method and a device for MCMC framework-based sub-hypergraph matching are provided. Matching of object features is performed by constructing sub-hypergraphs. In a large number of actual images and videos, objects vary constantly, and contain various noise points as well as other interference factors, which makes image object matching and searching very difficult. Perform object feature matching by representing the appearance and positions of objects by sub-hypergraphs allows for faster and more accurate image matching. Furthermore, a sub-hypergraph has several advantages over a graph or a hypergraph: on one hand, a sub-hypergraph has more geometric information (e.g. angle transformation, rotation, scale, etc.) than a graph, and has a lower degree of difficulty and better extensibility than a hypergraph. On the other hand, the disclosed method and device have stronger capabilities to resist interference and good robustness, and are adaptable to more complex settings, especially with outliers.

MCMC framework-based sub-hypergraph matching method and device

The present application provides methods, systems, devices and computer-readable mediums for encoding and decoding transform. A method of the present application comprises: performing transform-encoding on a prediction residual block separately using a plurality of transform matrices, to obtain transformed residual blocks; based on the transformed residual block, determining, by a rate-distortion optimization decision, a transform matrix or a combination of transform matrices matching the residual characteristics of the prediction residual block from the plurality of transform matrices, and determining a transformed residual block to be outputted in a bitstream. Compared with the prior art, the method of the embodiments of the present invention performs a residual transform using transform matrices that are more closely matched with the residual characteristic, thereby improving the expression of the residual signal and improving the coding efficiency of the residual block.

Methods, systems, devices and computer-readable mediums for encoding and decoding transform matrices matching residual characteristics of prediction residual block

The present application provides a method and a device of encoding and decoding based on free viewpoint, and relates to the technical field of video encoding. The method includes: generating a planar splicing image and splice information based on multiple single-viewpoint videos at a server side; generating a planar splicing video based on the planar splicing image; generating camera side information of the planar splicing video based on camera side information existing in the multiple single-viewpoint videos; and encoding the planar splicing video, the splice information and the camera side information of the planar splicing video to generate a planar splicing video bit stream, and decoding a planar splicing video bit stream to acquire a virtual viewpoint according to viewpoint information of a viewer at client side.

Method and device of encoding and decoding based on free viewpoint

Disclosed are a panoramic video asymmetrical mapping method and a corresponding inverse mapping method that include mapping a spherical surface corresponding to a panoramic image or video A onto a two-dimensional image or video B, projecting the spherical surface onto an isosceles quadrangular pyramid with a square bottom plane, and further projecting the isosceles quadrangular pyramid onto a planar surface, using isometric projection on a main viewpoint region in the projection and using a relatively high sampling density to ensure that the video quality of the region of the main viewpoint is high, while using a relatively low sample density for non-main viewpoint regions so as to reduce bit rate. The panoramic video asymmetrical inverse mapping technique provides a method for mapping from a planar surface to a spherical surface, and a planar surface video may be mapped back to a spherical surface for rendering and viewing.

Main viewpoint-based panoramic video mapping method

Growing community of inventors

Shenzhen, China

Zhenyu Wang