-
1.发明公开公开(公告)号：US20240152806A1
公开(公告)日：2024-05-09
申请号：US18388078
申请日：2023-11-08
Inventor： Hao WU , Cheng YANG
IPC: G06N20/00
CPC classification number: G06N20/00
Abstract: A solution for generating a negative sample pair of a contrastive learning model is provided. In one method, a first data segment is obtained from a first data sequence in a plurality of data sequences for training the contrastive learning model, and a second data segment is obtained from a second data sequence in the plurality of data sequences. A data frame is selected from a further data sequence than the second data sequence in the plurality of data sequences. A third data segment is generated based on the second data segment and the data frame. A negative sample pair for training the contrastive learning model is determined based on the first data segment and the third data segment. Therefore, richer semantic information can be introduced into negative sample pairs in terms of appearance, and further the accuracy of the contrastive learning model can be improved.
-
2.发明公开公开(公告)号：US20240346318A1
公开(公告)日：2024-10-17
申请号：US18504220
申请日：2023-11-08
Inventor： Hao WU , Cheng Yang
Abstract: A method, apparatus, device, and medium for managing a model based on a distance between samples. In one method, a basic sample for training a contrastive learning model and a plurality of negative samples associated with the basic sample is obtained; a sequence of the plurality of negative samples is generated based on distances between the plurality of negative samples and the basic sample; the sequence of the plurality of negative samples is divided into a first set of negative samples and a second set of negative samples; an update parameter for updating the contrastive learning model is determined based on the basic sample, the first set of negative samples and a first weight of the first set of negative samples, and the second set of negative samples and a second weight of the second set of negative samples, the first weight is greater than the second weight.
-
Patent Agency Ranking
公开(公告)号：US20240112299A1
公开(公告)日：2024-04-04
申请号：US18255473
申请日：2021-12-01
Applicant: BEIJING YOUZHUJU NETWORK TECHNOLOGY CO. LTD.
Inventor： Hao WU , Changhu WANG
CPC classification number: G06T3/0012 , G06T3/4007
Abstract: This disclosure relates to a video cropping method and apparatus, storage medium, and electronic device. The present disclosure method: acquiring an original video to be cropped; performing frame extraction processing on the original video to obtain a plurality of target video frames; determining, for each of the target video frames, a target candidate cropping box corresponding to the target video frame according to a main content in the target video frame; performing interpolation processing according to the target candidate cropping box corresponding to each of the target video frames to determine a target cropping box corresponding to each frame picture in the original video; and cropping the original video according to the target cropping box corresponding to the each frame picture.
-
公开(公告)号：US20240323479A1
公开(公告)日：2024-09-26
申请号：US18573383
申请日：2022-08-30
Inventor： Hao WU , Qinyuan ZHANG , Liangcheng XU
IPC: H04N21/431 , H04N21/482
CPC classification number: H04N21/4316 , H04N21/4826
Abstract: The present disclosure relates to a media display method, a media display apparatus, and a storage medium. The media display method includes providing a media display page, which includes a media source display region; obtaining respective identification information of currently available media sources and displaying the identification information of the currently available media sources in the media source display region; and in response to a first operation instruction for the media display page, changing a display manner of the media source display region.
-
公开(公告)号：US20240152816A1
公开(公告)日：2024-05-09
申请号：US18504931
申请日：2023-11-08
Inventor： Hao WU , Cheng YANG
IPC: G06N20/00
CPC classification number: G06N20/00
Abstract: A method, apparatus, device, and medium for managing a contrastive learning model are provided. In one method, in a first training phase, a first contrastive learning model is generated by training the contrastive learning model with a first training sample set, a negative sample pair of the first training sample set comprising only data segments from different data sequences. In a second training phase, a second contrastive learning model is generated by training the first contrastive learning model with a second training sample set, a negative sample pair in the second training sample set comprising data segments from the same data sequence. Knowledge in terms of the appearance of the samples can be fully obtained in the first training phase, and knowledge in terms of the appearance and dynamics of the samples can be fully obtained in the second training, with example implementations of the present disclosure.
-
6.发明公开公开(公告)号：US20240152815A1
公开(公告)日：2024-05-09
申请号：US18503658
申请日：2023-11-07
Inventor： Hao WU , Cheng YANG
IPC: G06N20/00
CPC classification number: G06N20/00
Abstract: A solution for generating a positive sample pair of a contrastive learning model is provided. In one method, a first data segment and a second data segment are respectively obtained from a first data sequence in plurality of data sequences for training the contrastive learning model, the first data sequence comprising a plurality of data frames. A data frame is selected from a second data sequence in the plurality of data sequences. A third data segment is generated based on the second data segment and the data frame. A positive sample pair for training the contrastive learning model is determined using the first data segment and the third data segment. In this way, positive samples that are more difficult to distinguish can be provided, thereby increasing the accuracy of the contrastive learning model and improving the training efficiency and accuracy of downstream models.
-
公开(公告)号：US20230394625A1
公开(公告)日：2023-12-07
申请号：US18253357
申请日：2021-11-16
Applicant: BEIJING YOUZHUJU NETWORK TECHNOLOGY CO. LTD.
Inventor： Hao WU , Yuntao MA , Changhu WANG
CPC classification number: G06T3/40 , G06T7/70 , G06V20/49 , G06V20/41 , G06V20/46 , G11B27/34 , G06T2207/10016 , G06T2207/30201
Abstract: The present disclosure relates to a video processing method and apparatus, a readable medium, and an electronic device, which relate to the technical field of image processing; the method includes: preprocessing a target video to obtain a plurality of target image frames of the target video; identifying a position of a designated object in each of the target image frames; and determining a reserved image frame from the target image frames based on the position of the designated object in each of the target image frames, the reserved image frame being used to indicate a cropping on image frames before the reserved image frame in the target video.
-
-
-
-
-
-
Search Results
7
records
(took 0.018 seconds)
