Model training method and model training device

Abstract

The model training method and training device provided by the embodiment of the invention are used for improving the simulation speed of the posture of the virtual digital human. The method comprises the steps that: a generator and a discriminator of an image translation model are trained through first data in a training set, and the trained image translation model is determined as a first image translation model, wherein the generator is of a coding model-decoding model structure, the coding model adopts a residual network architecture, the first data comprises a target frame image, contour line data of the target frame image, distance image data of the target frame image, first N frames of images of the target frame image, contour line data of the first N frames of images and distance image data, and N is an integer greater than or equal to 2; the target frame image is any one frame or any multiple frames of images except the first frame image and the second frame image in the training set; and the residual network architecture of the coding model of the first image translation model is modified into a lightweight model architecture to obtain a second image translation model.

Description

technical field [0001] The invention relates to the technical field of image translation, in particular to a model training method and a model training device. Background technique [0002] The so-called image translation refers to the transformation from one image to another. It can be compared to machine translation, converting one language to another. [0003] Typical image translation models in the prior art include pix2pix, pix2pixHD, and vid2vid. pix2pix proposes a unified framework to solve various image translation problems, pix2pixHD solves the problem of high-resolution image conversion (translation) better on the basis of pix2pix, and vid2vid better solves the problem of high-resolution image conversion (translation) on the basis of pix2pixHD Fixed video conversion issues with high resolutions. [0004] A digital human is a virtual simulation of the human body at different levels of form and function using the method of information science. The current image t...

AI Technical Summary

Problems solved by technology

The current image translation model can perform virtual simulation of the digital human pose in the image, but the image translation model in the prior art has a complex model structure, which leads to a large amount of data calculation in the training process, which in turn makes image translation difficult. The image translation speed of the model is slow, that is, the speed of digital human pose simulation is slow

Images