(S) Semantically Annotated Deformable, Soft and Rigid BodiesΒΆ
In the core of MAGES lies an advanced mathematical algorithm for physics-based visual techniques to allow the 3D representation of deformable soft body objects (skin, tissue, etc.), essential for VR surgical training.
Since surgical training is all about cutting and suturing soft body objects, collision detection (touching) and handling of soft bodies with other objects becomes crucial for high-realism VR. The soft deformation algorithm is based on shape matching techniques and particle-based soft body simulations.
Our methodology differs from the state of the art since it provides onthe-go control of the particles as physical objects and a centre point, which controls the entire soft body. Velocity based interaction can be applied directly to the corresponding particles while interacting with the environment as objects.
In addition we synchronize the deformable objects over network utilizing our GA interpolation engine to improve interaction with concurrent users.
Finally, we extended our physics engine to support interaction with ropes and even giving user the ability to perform knots and sutures in VR. In combination with our soft body mechanics we developed a bowel anastomosis operation where users can interact with the virtual sutures and soft tissue.