VolCo: Volumetric Contact for High-Fidelity Human Grasp Generation

Key points

• Accurate contact modeling is fundamental to understanding hand-object interaction, yet existing contact representations are typically restricted to object surfaces and rely on hand-crafted rules to recover contact details, leading to severe penetrations and implausible results.1
• VolCo encodes 3D contact that allows precise hand part recovery, and is organized in an inherent hierarchy: local contact details within each volume and global hand geometry across all volumes.1
• For local contact details, we use a 3D variational autoencoder to model the possible hand configurations conditioned on the local object signed distance field (SDF).1
• For global hand geometry, we design a prior-guided diffusion model that learns the distribution of compressed latent features aggregated from the volumetric grids.1

Sources (1)

• [1]VolCo: Volumetric Contact for High-Fidelity Human Grasp Generation
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 7, 02:58 PM

  To better exploit the rich detail in motion-capture data, we introduce Volumetric Contact (VolCo), a representation that expands surface points to a set of 3D volumetric grids.

  Accurate contact modeling is fundamental to understanding hand-object interaction, yet existing contact representations are typically restricted to object surfaces and rely on hand-crafted rules to recover contact details, leading to severe penetrations and implausible results.

Extractive summary: sentences quoted from the sources.