Make-It-Poseable: Feed-forward Latent Posing Model for 3D Humanoid Character Animation
This addresses robustness and generalizability issues in 3D character animation for computer graphics and vision applications, representing a novel method for a known bottleneck.
The paper tackles the problem of inaccurate skinning weight prediction, topological imperfections, and poor pose conformance in 3D character posing by introducing Make-It-Poseable, a feed-forward framework that reformulates posing as latent-space transformation, resulting in superior posing quality and natural extension to 3D editing applications.
Posing 3D characters is a fundamental task in computer graphics and vision. However, existing methods like auto-rigging and pose-conditioned generation often struggle with challenges such as inaccurate skinning weight prediction, topological imperfections, and poor pose conformance, limiting their robustness and generalizability. To overcome these limitations, we introduce Make-It-Poseable, a novel feed-forward framework that reformulates character posing as a latent-space transformation problem. Instead of deforming mesh vertices as in traditional pipelines, our method reconstructs the character in new poses by directly manipulating its latent representation. At the core of our method is a latent posing transformer that manipulates shape tokens based on skeletal motion. This process is facilitated by a dense pose representation for precise control. To ensure high-fidelity geometry and accommodate topological changes, we also introduce a latent-space supervision strategy and an adaptive completion module. Our method demonstrates superior performance in posing quality. It also naturally extends to 3D editing applications like part replacement and refinement.