Face image fusion method and device, storage medium and electronic equipment

Abstract

The invention provides a face image fusion method and device, a storage medium and electronic equipment, and the method comprises the steps of obtaining a face template image and a to-be-fused face image corresponding to a face fusion instruction when the face fusion instruction is received; performing multi-level down-sampling on the face template image to obtain original attribute features of the face template image, and performing multi-level up-sampling on the original features to obtain attribute features generated by each level of up-sampling; performing feature extraction on the to-be-fused face image by using a face recognition model to obtain an identity feature of the to-be-fused face image; and inputting the original attribute features, the attribute features generated by each level of up-sampling and the identity features into a pre-constructed feature fusion model to obtain a face fusion image output by the feature fusion model. By applying the method provided by the embodiment of the invention, the image fusion effect can be improved under the condition that the face poses and visual angles of the to-be-fused source image and the template image are greatly different.

Description

technical field [0001] The present invention relates to the technical field of data processing, in particular to a face image fusion method and device, a storage medium and electronic equipment. Background technique [0002] With the development of society and science and technology, people's demand for good portrait shooting effects is increasing day by day. Functions such as face beautification, stickers, hairstyle change, and face change in various camera applications or image processing applications are widely welcomed by users. Among them, face changing is the fusion of face images, which mainly combines user photos and material photos, so that the fused image has both the facial appearance features in the user photos and the character image features in the material photos. This meets the user's entertainment needs and improves the fun of the application. [0003] In the current image processing technology, the pixels in the face area are usually directly replaced to ...

AI Technical Summary

Problems solved by technology

[0003] In the current image processing technology, the pixels in the face area are usually directly replaced to realize the fusion of the user's face image and the face template image, but this method is very sensitive to the face pose and angle of view. When the user's source image to be fused and the face pose and perspective of the template image are quite different, the effect of image fusion is poor.

Images