Human body modeling model acquisition method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a human body modeling model acquisition method and device, electronic equipment and a storage medium. The key point loss of the true value key point information and the predicted key point information is calculated, and the loss of the deep learning model is calculated and obtained in combination with the key point loss and the human body mask loss, therebyreversely adjusting the parameters of the deep learning model according to the loss of the deep learning model. When the loss of the deep learning model is calculated, key point loss is added, the influence of key points on parameter adjustment can be effectively improved, and therefore the performance of the deep learning model is improved. When the loss of the deep learning model is calculated,the human body masking loss is added, and the area where the human body is located, the human body shape and the like can be restrained.

Description

technical field [0001] The present invention relates to the technical field of deep learning, in particular to a method, device, electronic equipment and storage medium for acquiring a human body modeling model. Background technique [0002] With the development of computer vision technology, especially the emergence of deep learning algorithms, image-based artificial intelligence technology has risen rapidly. 3D (Dimension, dimension) human body reconstruction is a branch algorithm in the field of computer vision, which can be used in applications such as human body intention recognition, virtual image establishment, human motion capture, human interactive games, etc., using human body images to obtain 3D human body masks, specifically as follows Figure 1a-Figure 1d shown, where Figure 1a for the human body image, Figure 1b A body mask for the main view of the 3D mannequin, Figure 1c A body mask for the left view of the 3D mannequin, Figure 1d Human mask for the rig...

AI Technical Summary

Problems solved by technology

[0004] However, the inventor found in the research that the importance of each point in the human body mask to the modeling of the human body mask is different. The importance of mask modeling, so using the above method, the performance of the trained deep learning model is poor

Images