DAMA: Disentangled Body-Anchored Gaussians for Controllable Multi-Layered Avatars

2026-05-20

AI Summary

Existing methods for 3D clothed human reconstruction often neglect geometric structure and physical plausibility, making it difficult to achieve clear garment layering and controllable stacking. This work proposes a multi-view reconstruction framework based on Gaussian representations, which explicitly models multi-layer garments with well-defined layer ordering and physical realism by anchoring Gaussians to the SMPL-X mesh surface and integrating 2D segmentation cues. The approach introduces voxel-anchored Gaussians, barycentric coordinate parameterization, topology-guided layer optimization, and joint geometry-appearance refinement, enabling—for the first time—a layered clothing representation that supports user-defined reordering and clean separation. Evaluated on the 4D-DRESS dataset, the method achieves state-of-the-art performance in geometric accuracy, garment separation quality, interpenetration rate, and penetration depth, while efficiently producing simulation-ready meshes.

Abstract

Existing 3D clothed avatar reconstruction methods achieve high visual fidelity but ignore geometric structure and physical plausibility. They either model clothed humans as a single deformable surface or attempt garment disentanglement without enforcing geometric constraints, resulting in ambiguous garment boundaries and no control over stacking or layer ordering. To address these limitations, we introduce DAMA (Disentangled body-Anchored Gaussians for Controllable Multi-layered Avatars), a 3D avatar reconstruction method that produces physically plausible clothed avatars through a dedicated representation and reconstruction method. At the representation level, we bind Gaussians to SMPL-X faces using barycentric in-plane coordinates and a positive normal offset. Based on this parameterization, the reconstruction method lifts 2D segmentations to body-anchored Gaussians, refines layers using topology-guided correction, and jointly optimizes geometry and appearance. DAMA is the first Gaussian avatar reconstruction method from multi-view images to achieve physically plausible layering, clean garment separation, and explicit stacking control. On the full 4D-DRESS dataset (82 scans), it achieves state-of-the-art performance in geometry reconstruction, garment separation, penetration rate, and penetration depth. The representation further supports user-defined garment reordering and fast conversion of body-conforming garments to simulation-ready meshes. Project Page: https://danieleskandar.github.io/dama/

Problem

• 3D avatar reconstruction
• garment disentanglement
• physical plausibility
• layer ordering
• geometric constraints

Innovation

• Disentangled Gaussians
• Body-Anchored Representation
• Multi-layered Avatars
• Physically Plausible Layering
• Topology-Guided Correction