Saved in:
Bibliographic Details
Main Authors: He, Yumeng, Jiang, Ying, Lu, Jiayin, Yang, Yin, Jiang, Chenfanfu
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.01629
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918227302416384
author He, Yumeng
Jiang, Ying
Lu, Jiayin
Yang, Yin
Jiang, Chenfanfu
author_facet He, Yumeng
Jiang, Ying
Lu, Jiayin
Yang, Yin
Jiang, Chenfanfu
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2512_01629
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SPARK: Sim-ready Part-level Articulated Reconstruction with VLM Knowledge
He, Yumeng
Jiang, Ying
Lu, Jiayin
Yang, Yin
Jiang, Chenfanfu
Computer Vision and Pattern Recognition
Robotics
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.
title SPARK: Sim-ready Part-level Articulated Reconstruction with VLM Knowledge
topic Computer Vision and Pattern Recognition
Robotics
url https://arxiv.org/abs/2512.01629