Model training method, electronic equipment and storage medium

Abstract

The embodiment of the invention relates to the field of data processing, and discloses a model training method, electronic equipment and a storage medium. In some embodiments of the invention, the model training method comprises the following steps: obtaining an image training set and annotation data, wherein the annotation data comprises a human hand three-dimensional model obtained based on image training set optimization; and based on a predefined loss function at least comprising adjacent frame loss constraint terms, training by using each frame of training image in the image training set and the annotation data to obtain a hand capture model, wherein the human hand capturing model is a neural network model used for judging human hand posture parameters and human hand shape parameters represented by the single human hand image, so that a human hand action model is obtained based on the human hand posture parameters, the human hand shape parameters and a human hand parameterized model. According to the embodiment, the parameters required in the process of constructing the human hand action model based on the parameterized model can be obtained based on the single image.

Description

technical field [0001] The embodiments of the present invention relate to the field of data processing, and in particular to a model training method, electronic equipment, and a storage medium. Background technique [0002] The reconstruction and attribute recognition of 3D human hands has always been an important research direction in the field of machine vision. At present, the work related to human reconstruction based on deep learning in academia can be roughly divided into two categories, parametric model reconstruction and non-parametric model reconstruction. Among them, non-parametric model reconstruction mainly takes multi-view depth maps and color maps as input, and obtains human hand models through a large amount of data learning, which is difficult to obtain data. Parametric model reconstruction is mainly to learn the parameters of the human hand model, and then use the parameters to fit the target gesture model. Both approaches have pros and cons. [0003] Wit...

AI Technical Summary

Problems solved by technology

Among them, non-parametric model reconstruction mainly takes multi-view depth maps and color maps as input, and obtains human hand models through a large amount of data learning, which is difficult to obtain data.

Images