Viet-Anh on Software Logo

What is: imGHUM?

SourceimGHUM: Implicit Generative Models of 3D Human Shape and Articulated Pose
Year2000
Data SourceCC BY-SA - https://paperswithcode.com

imGHUM is a generative model of 3D human shape and articulated pose, represented as a signed distance function. The full body is modeled implicitly as a function zero-level-set and without the use of an explicit template mesh. We compute the signed distance s=S(ρ,α)s = S\left(\rho, \alpha\right) and the semantics c=C(ρ,α)c = C\left(\rho, \alpha\right) of a spatial point ρ\rho to the surface of an articulated human shape defined by the generative latent code α\alpha. Using an explicit skeleton, we transform the point ρ\rho into the normalized coordinate frames as {ρ~j\tilde{\rho}^{j}} for N=4N = 4 sub-part networks, modeling body, hands, and head. Each sub-model {SjS^{j}} represents a semantic signed-distance function. The sub-models are finally combined consistently using an MLP U to compute the outputs ss and cc for the full body. The multi-part pipeline builds a full body model as well as sub-part models for head and hands, jointly, in a consistent training loop.

On the right of the Figure, we visualize the zero-level-set body surface extracted with marching cubes and the implicit correspondences to a canonical instance given by the output semantics. The semantics allows e.g. for surface coloring or texturing.