GRADRobot: Geometry-Aware Rendering with Articulation and Diffusion for Robot Modeling

Yunlong Li, Boyuan Chen, Chongjie Ye, Bohan Li, Zhaoxi Chen, Shaocong Xu, Hao Tang, Hao Zhao

Keywords: Gaussian Splatting, Articulated Rendering, Diffusion Refinement

TL;DR: GRADRobot is a geometry–diffusion hybrid renderer that combines a surface-anchored canonical Gaussian field with pose-conditioned deformation and diffusion-based refinement Abstract:

Abstract: Gaussian fields are a promising representation for robot body modeling due to their differentiability and inherently low sim-to-real gap. However, existing methods like DrRobot overlook explicit geometric constraints, leading to artifacts under novel poses or views. Directly enforcing depth and normal supervision on articulated Gaussians is unstable due to entanglement between pose deformation and 3D appearance learning. To address this, we propose a two-stage training strategy: we first learn a canonical Gaussian field in a canonical pose using dense RGB, depth, and normal supervision, establishing a geometry-aware reconstruction. We then fine-tune the Gaussian parameters jointly with a deformation network conditioned on joint angles using only RGB losses, ensuring consistent geometry and appearance across poses. To further mitigate rendering artifacts in novel poses and viewpoints, we integrate a diffusion-based refinement module. This module conditions on both the initial Gaussian renderings and the target robot skeletons, and significantly enhances visual fidelity while preserving pose accuracy. Experiments across multiple robotic platforms show that GRADRobot outperforms DrRobot by a large margin in both rendering quality (PSNR) and geometric accuracy (Chamfer Distance).

Serve As Reviewer: ~Hao_Zhao1

Submission Number: 7