-
公开(公告)号：EP4462771A2
公开(公告)日：2024-11-13
申请号：EP23737083.8
申请日：2023-01-05
发明人： CHEN, Fangdong , PAN, Dongping
IPC分类号： H04N19/103
摘要： The present application discloses image coding and decoding methods and apparatuses, and a storage medium, relates to the field of image coding and decoding technologies, and helps to improve coding and decoding efficiency. An image decoding method includes: parsing a code stream to obtain a first syntax element, where the first syntax element includes an index of a target prediction mode of a to-be-decoded unit; determining the target prediction mode from an index table based on the index of the target prediction mode, where the index table includes correspondences between indexes of multiple prediction modes and the multiple prediction modes; reconstructing the to-be-decoded unit based on at least the target prediction mode to obtain a reconstructed block.
-
公开(公告)号：EP4460007A1
公开(公告)日：2024-11-06
申请号：EP22914332.6
申请日：2022-12-16
发明人： WANG, Hou
IPC分类号： H04N21/234
摘要： Embodiments of the present application provide a video transcoding method, a video display method, an apparatus and an electronic device, which relate to the field of video processing technology. The video transcoding method includes: acquiring a first image in an initial images obtained by stitching multi-channel video images; determining an initial resolution of each of initial sub-images; wherein each initial resolution is not greater than a maximum resolution of input data that can be supported by a preset encoder, and a target resolution of a target sub-image obtained after transcoding each initial sub-image is not greater than a maximum resolution of output data that can be supported by the preset encoder; segmenting a to-be-cut image into respective initial sub-images based on the respective initial resolutions; encoding each initial sub-image according to a target resolution of a target sub-image after transcoding this initial sub-image by using a preset encoder, to obtain each of target code-streams. Compared with the prior art, by applying the solution provided by the embodiments of the present application, it is possible to transcode the video screen obtained after stitching multi-channel video screens.
-
公开(公告)号：EP4435660A1
公开(公告)日：2024-09-25
申请号：EP22894711.5
申请日：2022-11-11
发明人： CHEN, Binbin , CHEN, Weijie , YANG, Shicai
IPC分类号： G06F30/27
摘要： The present disclosure provides an object detection method and apparatus, including: adding pseudo labels to unlabeled data, dividing pseudo labels into high-quality pseudo labels and uncertain pseudo labels; inputting the unlabeled data to an initial learning model to obtain first predicted values; determining first predicted labels and first predicted boxes based on first predicted values corresponding to the high-quality pseudo labels, and determining second predicted labels and second predicted boxes based on first predicted values corresponding to the uncertain pseudo labels; inputting the unlabeled data to an initial management model to obtain second predicted values, and determining third predicted labels and third predicted boxes based on second predicted values corresponding to the uncertain pseudo labels; training the initial management model based on the first predicted labels, first predicted boxes, second predicted labels, second predicted boxes, third predicted labels and third predicted boxes to obtain a target management model, which is used to perform object detection on the data to be detected.
-
公开(公告)号：EP4277267A1
公开(公告)日：2023-11-15
申请号：EP22758868.8
申请日：2022-02-22
发明人： CHEN, Fangdong , CAO, Xiaoqiang , SUN, Yucheng
IPC分类号： H04N19/117
摘要： In the present disclosure, an encoding and decoding method and apparatus and devices thereof are provided. The method may include: if a current pixel point in a current block satisfies an enable condition of an enhanced refinement mode, determining a gradient value of the current pixel point based on an original pixel value of the current pixel point and an original pixel value of a surrounding pixel point of the current pixel point; based on the gradient value of the current pixel point and the original pixel value of the current pixel point, determining a refined pixel value of the current pixel point. The technical solutions of the present disclosure can improve the encoding performance.
-
5.发明公开
公开(公告)号：EP4266244A1
公开(公告)日：2023-10-25
申请号：EP21905842.7
申请日：2021-12-17
发明人： ZHANG, Yingying , ZHONG, Qiaoyong , XIE, Di , PU, Shiliang
IPC分类号： G06T7/00
摘要： Embodiments of the present disclosure discloses a surface defect detection method, device, system, storage medium, and program product, belonging to the field of deep learning technology. In the embodiments of the present disclosure, if it is detected and determined through a neural network defect segmentation model that there are no surface defects of known types in the to-be-detected image, the image features of the to-be-0+22 image is extracted through the neural network defect feature extraction model, to obtain the to-be-compared features. If the similarity between the to-be-compared features and the normal data representation features is small, it is determined that there is a surface defect of an unknown type in the to-be-detected image, that is, the embodiments of the present disclosure can detect surface defects of unknown defect types, improving the accuracy of detection.
-
公开(公告)号：EP4243414A1
公开(公告)日：2023-09-13
申请号：EP22770372.5
申请日：2022-03-10
发明人： WEI, Liang , CHEN, Fangdong , WANG, Li
IPC分类号： H04N19/159
摘要： Disclosed are a decoding method based on an adaptive intra-frame refreshing mechanism, an encoding method based on an adaptive intra-frame refresh mechanism, and a related device, which belong to the field of video encoding and decoding. In the embodiments of the present application, an adaptive intra-frame video refreshing extension number and virtual boundary location mark information are newly added to extended data in a code stream of the current frame. In this way, since virtual boundary location mark information can indicate the location of a virtual boundary, and the virtual boundary is at least used for distinguishing between a refreshed area and a non-refreshed area in the current frame, when dividing the refreshed area of the current frame, an encoding terminal can arbitrarily set the virtual boundary on the basis of actual demands, and indicate the location of the virtual boundary by means of the virtual boundary location mark information, thereby effectively improving the flexibility of area division.
-
公开(公告)号：EP4198627A1
公开(公告)日：2023-06-21
申请号：EP21913104.2
申请日：2021-08-11
发明人： YE, Beibei , LIANG, Xue
摘要： The present disclosure provides a light supplementing lamp for a hemispherical camera and the hemispherical camera. The light supplementing lamp for the hemispherical camera includes a lamp source configured for generating a light supplementing source; a reflecting element configured for reflecting a light supplementing source generated by the lamp source; a lamp cover configured for transmitting a light supplementing source reflected by the reflecting element, so that a light supplementing area of the light supplementing lamp is suitable for a video monitoring area of the hemispherical camera; wherein, the reflecting element and the lamp cover are respectively arranged at two sides of the lamp source, and the reflecting element is arranged opposite to the lamp source.
-
8.发明公开
公开(公告)号：EP4187877A1
公开(公告)日：2023-05-31
申请号：EP21845873.5
申请日：2021-07-20
发明人： LI, Zhixiao
摘要： Embodiments of the present application relate to the technical field of communications. Disclosed are a method and apparatus for establishing communication connection, and a device, and a computer readable storage medium. The method comprises: establishing communication connection with a server on the basis of a first port, the first port being a port through which a client negotiates with the server, or, the first port being a port used when the client communicates with the server last time; in the case that establishing the communication connection with the server on the basis of the first port fails, reading port information in target configuration data, a second port indicated by the port information in the target configuration data being used for establishing the communication connection between the client and the server; and establishing the communication connection with the server on the basis of the second port. The client does not need to traverse a plurality of ports but can directly determine, by reading the target configuration data, the second port that can be used for establishing the communication connection with the server, and therefore, the time required for establishment of the communication connection between the client and the server can be shortened, thereby improving the communication efficiency.
-
公开(公告)号：EP4145827A1
公开(公告)日：2023-03-08
申请号：EP21816692.4
申请日：2021-05-31
发明人： SUN, Yucheng
IPC分类号： H04N19/105 , H04N19/139
摘要： Encoding and decoding methods and apparatuses, coder-side devices and decoder-side devices are provided, the method includes: obtaining a weighted prediction angle and weight configuration parameters of the current block; where the weight configuration parameters include a weight transform rate and a weight transform start position; configuring reference weight values for surrounding positions outside the current block according to the weight configuration parameters; determining a surrounding matching position to which the pixel position points from the surrounding positions outside the current block based on the weighted prediction angle; determining a target weight value of the pixel position based on the reference weight value associated with the surrounding matching position, determining an association weight value of the pixel position based on the target weight value of the pixel position; determining a first prediction value of the pixel position based on a first prediction mode, determining a second prediction value of the pixel position based on a second prediction mode; determining a weighted prediction value of the pixel position based on the first prediction value, the target weight value, the second prediction value and the association weight value.
-
公开(公告)号：EP4131871A1
公开(公告)日：2023-02-08
申请号：EP21776884.5
申请日：2021-03-23
发明人： FU, Yanyun , ZHANG, Chao
IPC分类号： H04L12/751 , H04L12/44
摘要： The present application discloses a method for generating a network topology, which determines multiple designated devices in a network, wherein the multiple designated devices are directly or indirectly connected through ports. The accessible devices corresponding to the ports of the designated devices among the multiple designated devices are determined, wherein any accessible device corresponding to a port of a designated device is directly or indirectly connected to the designated device through the port. A root node device among the multiple designated devices is determined according to the number of accessible devices corresponding to the ports of the designated device(s). The designated devices that are directly connected among the multiple designated devices are determined according to the accessible devices corresponding to the ports of the designated devices. The topology of the multiple designated devices in the network is generated according to the root node device and the designated devices that are directly connected among the multiple designated devices. Thus, the topology of multiple designated devices in the network is automatically generated according to the accessible devices corresponding to the ports of the multiple designated devices in the network, which is simple, efficient and accurate.
-
-
-
-
-
-
-
-
-
搜索结果
341 条记录 (耗时 0.031 秒)
