Face attribute editing model training method, face attribute editing method and equipment

Abstract

The embodiment of the invention discloses a training method of a face attribute editing model, a face attribute editing method and equipment, and belongs to the field of artificial intelligence. The method comprises the following steps: inputting a first picture into a picture encoder to obtain encoding features output by the picture encoder; respectively inputting the coding features into a first picture decoder and a second picture decoder to obtain a second picture and a third picture; based on the second picture and the third picture, a target loss function of a first picture decoder is constructed, the target loss function comprises adversarial loss and feature matching loss, and the feature matching loss is used for constraining similarity of deep semantic features between the pictures; and training a first picture decoder based on the target loss function, and determining the picture encoder and the trained first picture decoder as a face attribute editing model. By adopting the scheme provided by the embodiment of the invention, the consistency of deep features in the generated picture and the original picture can be ensured, and the editing quality of the face attribute can be improved.

Description

technical field [0001] The embodiments of the present application relate to the field of artificial intelligence, and in particular to a training method of a face attribute editing model, a face attribute editing method and equipment. Background technique [0002] Face attribute editing is a technology that changes specific attributes of faces in pictures or videos. For example, through face attribute editing, it is possible to make faces in pictures look older or younger, change male faces into female faces, or Become a star face. [0003] In related technologies, a face attribute editing model is usually obtained based on auto-encoder training for face attribute editing. Among them, when training the face attribute editing model, the adversarial loss is usually used as the loss function for model training. [0004] However, when using the adversarial loss as the loss function for model training, since the training process only focuses on the attributes to be edited, it i...

AI Technical Summary

Problems solved by technology

[0004] However, when using the adversarial loss as the loss function for model training, since only the attributes to be edited are focused on during the training process, it is easy to cause the loss of the features of the generated image and affect the quality of face attribute editing.

Images