Proxy Mesh Rendering
Rendering depth maps, normal maps or segmentation masks from coarse parametric body or head meshes according to target camera and pose, used as spatial conditioning input for diffusion models to provide a geometric scaffold.
Timeline (last 90 days)
VDAs delegate control of camera viewpoint and upper-body pose uniformly to proxy mesh rendering, while expressions go through the learned encoding channel, so the three are independent of one another and can be adjusted separately.
Proxy mesh rendering refers to rendering depth maps, normal maps, or segmentation masks from coarse parametric human/head meshes according to the target camera and pose, as spatial conditioning inputs to diffusion models.
All Facts (2)
VDAs delegate control of camera viewpoint and upper-body pose uniformly to proxy mesh rendering, while expressions go through the learned encoding channel, so the three are independent of one another and can be adjusted separately.
50%UnverifiedProxy mesh rendering refers to rendering depth maps, normal maps, or segmentation masks from coarse parametric human/head meshes according to the target camera and pose, as spatial conditioning inputs to diffusion models.
50%