Human body three-dimensional model establishing method and device, electronic equipment and storage medium

Abstract

The invention relates to a human body three-dimensional model establishing method and device, electronic equipment and a storage medium. The precision of an established human body three-dimensional model can be improved. The method comprises the following steps: acquiring a to-be-processed image including a human body; inputting the to-be-processed image into the neural network model to respectively obtain a first intermediate feature map and the position of a two-dimensional key point of a human body in the to-be-processed image, wherein the first intermediate feature map is used for representing surface structure information of a human body in the to-be-processed image; performing fusion processing on the first intermediate feature map and the two-dimensional key point position of the human body to obtain a fused feature map; inputting the fused feature graph into a graph convolutional neural network to obtain pose parameters of the human body; inputting the fused feature map into a parameter extraction network to obtain human body form parameters and camera parameters; and establishing a human body three-dimensional model of the human body according to the pose parameters, the human body shape parameters and the camera parameters.

Description

technical field [0001] The present disclosure relates to the technical fields of computer vision and computer graphics, and in particular to a method, device, electronic equipment and storage medium for establishing a three-dimensional model of a human body. Background technique [0002] The establishment of a three-dimensional model of the human body is a key issue in the technical fields of computer vision and computer graphics. High-quality 3D human body models have broad application prospects and important application values ​​in the fields of film and television entertainment, demographic data analysis and other fields. [0003] At present, the establishment of high-quality 3D human body models usually relies on expensive laser scanners or multi-camera array systems. This method of establishing 3D human body models has poor versatility and limited applicable scenarios, and the modeling takes a long time. Using a deep neural network to reconstruct a 3D human body model ...

AI Technical Summary

Problems solved by technology

[0004] The present disclosure provides a method, device, electronic device, and storage medium for establishing a three-dimensional model of a human body, so as to solve the problem of low precision of a three-dimensional human body model established from a single image by using a deep neural network

Images