SPARK: Sim-ready Part-level Articulated Reconstruction with VLM Knowledge

TL;DR

SPARK is a novel framework that reconstructs detailed, simulation-ready articulated 3D objects from a single RGB image by combining vision-language models, generative diffusion, and differentiable optimization.

Contribution

It introduces a comprehensive pipeline that automates the creation of articulated 3D assets from images, integrating VLMs, diffusion transformers, and differentiable rendering for accurate, ready-to-use models.

Findings

Produces high-quality, simulation-ready articulated assets.

Effective across diverse object categories.

Enables downstream robotics applications.

Abstract

Articulated 3D objects are critical for embodied AI, robotics, and interactive scene understanding, yet creating simulation-ready assets remains labor-intensive and requires expert modeling of part hierarchies and motion structures. We introduce SPARK, a framework for reconstructing physically consistent, kinematic part-level articulated objects from a single RGB image. Given an input image, we first leverage VLMs to extract coarse URDF parameters and generate part-level reference images. We then integrate the part-image guidance and the inferred structure graph into a generative diffusion transformer to synthesize consistent part and complete shapes of articulated objects. To further refine the URDF parameters, we incorporate differentiable forward kinematics and differentiable rendering to optimize joint types, axes, and origins under VLM-generated open-state supervision. Extensive experiments show that SPARK produces high-quality, simulation-ready articulated assets across diverse categories, enabling downstream applications such as robotic manipulation and interaction modeling. Project page: https://heyumeng.com/SPARK/index.html.