公开(公告)号：US20240404018A1

公开(公告)日：2024-12-05

申请号：US18800385

申请日：2024-08-12

Applicant: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED

Inventor： Wenhui HAN ,  Yandan Zhao ,  Ying Tai ,  Donghao Luo ,  Chengjie Wang

IPC: G06T5/60 ,  G06T5/77 ,  G06T7/00 ,  G06V10/774 ,  G06V10/776 ,  G06V10/82 ,  G06V40/16

Abstract: An image processing method, performed by a computer device, includes: obtaining an input image to be processed; performing face detection on the input image to obtain an input face image to be processed including at least one defect element relating to a skin element; and inputting the input face image into an image processing model to obtain a target face image corresponding to the input face image without a first defect element amongst the at least one defect element, wherein a training sample of the image processing model includes a first face image with a first face distortion degree less than a preset threshold and that is annotated with the first defect element.

公开(公告)号：US11928893B2

公开(公告)日：2024-03-12

申请号：US17530428

申请日：2021-11-18

Applicant: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED

Inventor： Donghao Luo ,  Yabiao Wang ,  Chenyang Guo ,  Boyuan Deng ,  Chengjie Wang ,  Jilin Li ,  Feiyue Huang ,  Yongjian Wu

IPC: G06V20/40 ,  G06F18/213 ,  G06T7/246 ,  G06V40/20 ,  G06N3/02

CPC classification number: G06V40/20 ,  G06F18/213 ,  G06T7/246 ,  G06N3/02 ,  G06T2207/20081

Abstract: An action recognition method includes: obtaining original feature submaps of each of temporal frames on a plurality of convolutional channels by using a multi-channel convolutional layer; calculating, by using each of the temporal frames as a target temporal frame, motion information weights of the target temporal frame on the convolutional channels according to original feature submaps of the target temporal frame and original feature submaps of a next temporal frame, and obtaining motion information feature maps of the target temporal frame on the convolutional channels according to the motion information weights; performing temporal convolution on the motion information feature maps of the target temporal frame to obtain temporal motion feature maps of the target temporal frame; and recognizing an action type of a moving object in image data of the target temporal frame according to the temporal motion feature maps of the target temporal frame on the convolutional channels.