Face Key Point Offset Adjustment for Expression Texture Transfer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing methods fail to change facial expressions while preserving the original facial features of an image, leading to loss of original characteristics during face swapping.
Innovation Solution
An image processing method that adjusts face key points of a target face image based on user-provided face key point offset information, allowing for the attachment of an expression texture image from a reference face to a to-be-processed image, thereby changing the expression while maintaining the original facial features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If face A in an image is directly replaced by face B from another image, then the facial information is changed, but the original facial features are lost
Solution Approach 1:
The patent segments the face replacement process into three distinct components: (1) the base face image retaining original facial features, (2) the expression texture image containing only expression information, and (3) the blended result. This segmentation allows selective transfer of expression while preserving original facial features, resolving the contradiction between changing facial information and maintaining feature accuracy.
Solution Approach 2:
The patent extracts the expression texture from the source face image by isolating only the expression-related information while removing the original facial features. This extracted expression texture is then applied to the target face, enabling expression transfer without feature loss. The extraction process solves the contradiction by separating expression from identity characteristics.
2Adaptability or versatility
If conventional face swapping is performed, then facial information changes, but the expression cannot be changed while reserving original facial features
Solution Approach 1:
The patent introduces an intermediary component called the 'expression texture image' that serves as a mediator between the source expression and the target face. This intermediary contains only expression information without the source face's identifying features, allowing expression transfer while maintaining the target face's original features. The intermediary resolves the contradiction by decoupling expression from identity.
Solution Approach 2:
The patent applies local quality by treating different regions of the face differently: the expression texture is applied locally to the facial region while the rest of the image maintains its original characteristics. This localized application ensures that only the expression changes while other facial features remain consistent and reliable.
Data Source
AI summary
An image processing method includes: obtaining a first target face image in a reference expression state; obtaining user face key point offset information that includes offsets between face key points of a user in a target expression state and face key points of the user in the reference expression state; adjusting face key points of the first target face image in the reference expression state according to the user face key point offset information; and attaching, in a to-be-processed image including a second target face image, an expression texture image of the adjusted first target face image to the second target face image, the second target face image and the first target face image belonging to a same target.