This series turns a static glTF character into a fully animated, physically-aware actor: compute-skinned on the GPU, ragdoll-capable, procedurally corrected, and expressive down to the face.
-
GPU compute skinning shared across rasterizer, ray tracing BLAS, and physics readback
-
Bone-proxy colliders, joint constraints, and animation-to-ragdoll handoff
-
Procedural animation: CCD/FABRIK inverse kinematics, foot placement, look-at, physics-driven lean
-
Bindless morph target buffers for facial animation at scale
-
A real production tooling and asset pipeline, not just a single demo scene
https://docs.vulkan.org/tutorial/latest/Advanced_glTF/introduction.html