-
31.发明公开
公开(公告)号：EP4105791A3
公开(公告)日：2023-03-15
申请号：EP22205704.4
申请日：2022-11-07
发明人： LYU, Shangwen , LI, Hongyu , LIU, Jing , WU, Hua , WANG, Haifeng
IPC分类号： G06F16/332 , G06N3/00 , G06N3/08
摘要： A method for training a document reading comprehension model includes: acquiring a question sample and a rich-text document sample, in which the rich-text document sample includes a real answer of the question sample; acquiring text information and layout information of the rich-text document sample by performing OCR processing on image information of the rich-text document sample; acquiring a predicted answer of the question sample by inputting the text information, the layout information and the image information of the rich-text document sample into a preset reading comprehension model; and training the reading comprehension model based on the real answer and the predicted answer. The method may enhance comprehension ability of the reading comprehension model to the long rich-text document, and save labor cost.
-
公开(公告)号：EP4079606A3
公开(公告)日：2023-03-15
申请号：EP22186142.0
申请日：2022-07-21
发明人： LI, Xinjing , WEI, Pengfei , JIAN, Huiling , SONG, Jia
摘要： A method and an apparatus for parking a vehicle and a medium are provided. The method comprises: acquiring parking environment data of a parking environment related to a vehicle and historical parking data in the parking environment; selecting, according to the parking environment data and the historical parking data, a constraint set for parking the vehicle from a plurality of constraint sets, the plurality of constraint sets corresponding to a corresponding parking risk level; and controlling the vehicle to park according to the selected constraint set. According to the solution of the present disclosure, the usage mode of the parking can be adaptively controlled for a risk level of the parking environment, thereby expanding the application scenarios of the parking technology.
-
33.发明公开
公开(公告)号：EP4145844A1
公开(公告)日：2023-03-08
申请号：EP22196149.3
申请日：2022-09-16
发明人： WANG, Xiangyang , XING, Huaifei
摘要： A method for detecting a jitter in a video includes obtaining (101) video frames of a video, in which the video frames include a target video frame and a plurality of historical video frames before the target video frame, determining (102) moving distances in a preset distance of the video frames relative to corresponding previous video frames, determining (103) a target amplitude and a target period in the preset direction for the target video frame; and determining (104) that there is a jitter in a video in response to the target amplitude in the preset direction determined for the target video frame being greater than a preset amplitude and the target period being less than a preset period.
-
公开(公告)号：EP4145298A1
公开(公告)日：2023-03-08
申请号：EP22197040.3
申请日：2022-09-22
摘要： With the method for snapshotting metadata, in response to reaching a current snapshot moment, a second basic version number of a binary search tree in a database at the current snapshot moment is generated according to a first basic version number at a previous snapshot moment; during a process from the current snapshot moment to a next snapshot moment, whenever metadata in the database is updated, the binary search tree is updated according to the updated metadata, and an updated version number of the binary search tree after each update is generated according to the second basic version number; and in response to reaching the next snapshot moment, a snapshot operation is performed on binary search trees corresponding to all version numbers generated between the current snapshot moment and the next snapshot moment to generate snapshot information of the current snapshot moment.
-
公开(公告)号：EP4140845A1
公开(公告)日：2023-03-01
申请号：EP22190582.1
申请日：2022-08-16
发明人： JIANG, Jianan , LIU, Xu , ZHOU, Jinyun , DONG, Fangfang , WAN, Guowei
摘要： The present disclosure provides a method and device for predicting a motion track of an obstacle and an autonomous vehicle, and relates to the technical field of autonomous driving, so as to at least solve the technical problem of low prediction precision of a motion track of an obstacle in an interaction scene. A specific implementation solution includes: environment information in a target scene, historical state information of a target obstacle and track planning information of a target vehicle are obtained, and the target obstacle is a potential interaction object of the target vehicle; and a motion track of the target obstacle is predicted based on the environment information, the historical state information and the track planning information.
-
公开(公告)号：EP4113387A3
公开(公告)日：2023-03-01
申请号：EP22192187.7
申请日：2022-08-25
发明人： SHI, Hongjian , FENG, Xinwei , LI, Feifei , GUO, Chenyang , WU, Xueqian , TIAN, Meng , SUN, Yu
IPC分类号： G06N3/042 , G06N3/09 , G06N3/045 , G06F16/9038
摘要： The present disclosure provides a search method based on a neural network model, the neural network model including a semantic representation model, a recall model, and a ranking model, and relates to the field of artificial intelligence, and in particular to the technical field of search. An implementation is: inputting a target search and a plurality of objects to be matched to the semantic representation model to obtain a first output of the semantic representation model, where the first output has a semantic understanding representation of recall and ranking; inputting the first output of the semantic representation model to the recall model, and obtaining at least one recall object matching the target search from the plurality of objects to be matched by using the recall model; and inputting a second output of the semantic representation model to the ranking model, and obtaining a matching value of each of the at least one recall object by using the ranking model, where the second output of the semantic representation model is obtained based on the input target search and the at least one recall object.
-
公开(公告)号：EP4113299A3
公开(公告)日：2023-02-22
申请号：EP22196817.5
申请日：2022-09-21
发明人： LIU, Chong , ZHUO, Zecheng , HUANG, Qiang , XU, Siqi , ZHANG, Wenjun , HONG, Saiding
摘要： A task processing method, a task processing device and an electronic device are provided, which relate to the field of cloud computing technology and big data technology, in particular to the field of task processing technology. The task processing method includes: obtaining a task processing request for a to-be-processed task, the task processing request including processing time information of the to-be-processed task and a service type of the to-be-processed task; in the case that the processing time information of the to-be-processed task meets a triggering condition, writing the to-be-processed task into a corresponding message queue in accordance with the service type of the to-be-processed task, one message queue corresponding to a respective one service type; and processing the to-be-processed task in the message queue, to obtain a task processing result of the to-be-processed task.
-
公开(公告)号：EP4134950A2
公开(公告)日：2023-02-15
申请号：EP22209126.6
申请日：2022-11-23
发明人： XIA, Shuai , HUANG, Weiqi , JIANG, Peng , TANG, Ruijian
IPC分类号： G10L15/22 , G06F3/16 , G10L15/183
摘要： The invention provides a method and an apparatus for speech-based instruction scheduling, and an electronic device, which relate to the field of speech technology. The method includes: a microphone included in an electronic device receiving a first input speech from a user; the electronic device converting the first input speech into a first text message; the electronic device obtaining first instruction contents of the first text message by parsing the first text message based on preset configuration instructions, in which the first instruction contents are configured to represent a target instruction object and a target instruction action; and the electronic device sending the first instruction contents to a scheduling object, so that the scheduling object controls the target instruction object to execute the target instruction action based on the first instruction contents. Therefore, it is achieved that the scheduling object is controlled based on a speech to perform automatic scheduling, and the scheduling efficiency is improved.
-
公开(公告)号：EP4131083A2
公开(公告)日：2023-02-08
申请号：EP22216303.2
申请日：2022-12-23
发明人： LIU, Ji , YU, Sunjie , DOU, Dejing , ZHOU, Jiwen
IPC分类号： G06N3/098 , G06N3/0464 , G06N3/082 , G06N3/084
摘要： A method for generating a federated learning model and an apparatus for generating a federated learning model are provided. The method includes obtaining images; obtaining sorting results of the images; and generating a trained federated learning model by training a federated learning model to be trained according to the images and the sorting results. The federated learning model to be trained is obtained after pruning a federated learning model to be pruned, and a pruning rate of a convolution layer in the federated learning model to be pruned is automatically adjusted according to a model accuracy during the pruning.
-
40.发明公开
公开(公告)号：EP4131024A1
公开(公告)日：2023-02-08
申请号：EP22191894.9
申请日：2022-08-24
发明人： GAN, Jingru , WANG, Haiwei , LUO, Jinchang , CHEN, Kunbin , HE, Wei , WANG, Shuhui
摘要： A method for extracting information, includes: obtaining an information stream comprising text and an image; generating, according to the text, embedded representations of textual entity mentions and a textual similarity matrix of the textual entity mentions and candidate textual entities; generating, according to the image, embedded representations of image entity mentions and an image similarity matrix of the image entity mentions and candidate image entities; and determining, based on an optimal transport, target textual entities of the textual entity mentions and target image entities of the image entity mentions according to the embedded representations of the textual entity mentions, the embedded representations of the image entity mentions, the textual similarity matrix and the image similarity matrix.
-
-
-
-
-
-
-
-
-
搜索结果
642 条记录 (耗时 0.022 秒)
