Deep learning-based facial appearance simulation driven by surgically planned craniomaxillofacial bony movement
A deep learning framework for predicting facial soft-tissue appearance from planned craniomaxillofacial bony movement.
← Back to Projects
Project
Facial simulation following bony movement for orthognathic surgical planning
Deep learning-based facial appearance simulation that estimates soft-tissue change from planned craniomaxillofacial bony movement.
- Developed ACMT-Net, an attentive correspondence-assisted movement transformation network with a CPSA module to estimate changes in one point set driven by the movement of another.
- Achieved accuracy comparable to finite-element simulation while substantially reducing computation time during planning.