VolumetricSMPL: A Neural Volumetric Body Model for Efficient Interactions, Contacts, and Collisions

2025-06-29

AI Summary

Traditional parametric human models (e.g., SMPL) rely on surface meshes, limiting efficient geometric interaction modeling between humans and scenes/objects; existing neural implicit volumetric models suffer from bottlenecks in joint robustness, computational efficiency, or memory overhead. This paper proposes VolumetricSMPL: a neural implicit volumetric representation integrating SMPL pose priors with differentiable signed distance functions (SDFs), augmented by Neural Blended Weights (NBW) that dynamically generate MLP parameters for shape- and pose-adaptive lightweight implicit modeling. The method enables high-fidelity contact modeling and self-intersection correction. Evaluated on human–object interaction reconstruction, scene-aware human recovery, and constrained motion synthesis, it significantly outperforms COAP—achieving 10× faster inference and 6× reduced GPU memory consumption—while maintaining superior accuracy, efficiency, and robustness.

Technology Category

• Humans and AI: Interaction Techniques and Devices
• Computer Vision: Diffusion Models for Vision
• Machine Learning: Learning with Manifolds

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Parametric human body models play a crucial role in computer graphics and vision, enabling applications ranging from human motion analysis to understanding human-environment interactions. Traditionally, these models use surface meshes, which pose challenges in efficiently handling interactions with other geometric entities, such as objects and scenes, typically represented as meshes or point clouds. To address this limitation, recent research has explored volumetric neural implicit body models. However, existing works are either insufficiently robust for complex human articulations or impose high computational and memory costs, limiting their widespread use. To this end, we introduce VolumetricSMPL, a neural volumetric body model that leverages Neural Blend Weights (NBW) to generate compact, yet efficient MLP decoders. Unlike prior approaches that rely on large MLPs, NBW dynamically blends a small set of learned weight matrices using predicted shape- and pose-dependent coefficients, significantly improving computational efficiency while preserving expressiveness. VolumetricSMPL outperforms prior volumetric occupancy model COAP with 10x faster inference, 6x lower GPU memory usage, enhanced accuracy, and a Signed Distance Function (SDF) for efficient and differentiable contact modeling. We demonstrate VolumetricSMPL's strengths across four challenging tasks: (1) reconstructing human-object interactions from in-the-wild images, (2) recovering human meshes in 3D scenes from egocentric views, (3) scene-constrained motion synthesis, and (4) resolving self-intersections. Our results highlight its broad applicability and significant performance and efficiency gains.

Problem

• Enhances human body model efficiency for interactions and collisions
• Reduces computational and memory costs in volumetric neural models
• Improves accuracy and speed in human-environment interaction tasks

Innovation

• Neural Blend Weights for efficient MLP decoders
• Dynamic weight matrices blending for computational efficiency
• Signed Distance Function for differentiable contact modeling