一种基于3D先验知识的人脸老化图像生成方法及系统
By employing a face aging method based on 3D prior knowledge, the shape and texture features of face images are decomposed and encoded. Combined with empirical mode decomposition and adversarial generative networks, the problem of generating low-quality aged images in existing technologies is solved, and high-quality and high-accuracy face aging image generation is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING UNIV OF POSTS & TELECOMM
- Filing Date
- 2023-08-22
- Publication Date
- 2026-07-17
AI Technical Summary
Existing facial aging methods cannot generate high-quality aging images, and the interaction between shape and texture attributes makes it difficult to accurately characterize the real process of facial aging.
A face aging image generation method based on 3D prior knowledge is adopted. The input face image is decomposed into UV location map and texture map, and then encoded and decoded respectively. Empirical mode decomposition and adversarial generative network are combined, and the network is optimized using multiple loss functions to generate high-quality aging images.
The generated aging images contain rich 3D information, and the shape and texture features are decoupled, achieving high-quality facial aging effects and improving the accuracy and realism of the images.
Smart Images
Figure CN117036559B_ABST