paper-with-me

Papers

PhyScene3D: Physically Consistent Interactive 3D Tabletop Scene Generation

2026-06-01 · Weixing Chen, Zhuoqian Feng, Yang Liu, Yexin Zhang, Yifan Wen, Yinghong Liao, Weichao Qiu, Guanbin Li, Liang Lin arxiv

Generating physically consistent 3D tabletop scenes is a fundamental yet underexplored problem for interactive and generalist robotic learning. The challenge stems from dense object hierarchies and irregular affordances. Here, an interactive scene denotes a physically valid, collision-free environment directly loadable into physics simulators. Existing methods, ranging from decoupled symbolic solvers to end-to-end regression models, often suffer from error propagation or overfitting to noisy supervision containing widespread physical violations. To address these limitations, we introduce PhyScene3D, a framework that reformulates generation as a Human-Mimetic Constructive Process. The proposed Cognitive Topological Reasoning Chain (CTRC) factorizes scene synthesis into a sequential, anchor-conditioned process. It employs a 3D AABB-based placement scheme that imposes a strong structural inductive bias. To address imperfect supervision and physical infeasibility, we introduce Physics-Aware Denoising Alignment (PADA). It integrates a differentiable Signed Distance Field (SDF) with Test-Time Optimization (TTO) to project generated scenes onto a physics-feasible manifold while preserving semantic intent. Experiments demonstrate that PhyScene3D outperforms state-of-the-art approaches in both semantic accuracy and physical validity, achieving a 40% reduction in scene-wise collision rate relative to the human-annotated training data.

📄 PDF Abstract BibTeX arXiv:2606.01649

Code (0)

등록된 구현이 없습니다.

Tasks

Scene Generation

Similar Papers 제목 키워드 기반

PhyScene: Physically Interactable 3D Scene Synthesis for Embodied AI

2024-04-15 · CVPR 2024 1 · Yandan Yang, Baoxiong Jia, Peiyuan Zhi, Siyuan Huang

With recent developments in Embodied Artificial Intelligence (EAI) research, there has been a growing demand for high-quality, large-scale interactive scene generation. While prior methods in scene synthesis have priorit…

Scene Generation

TabletopGen: Tabletop Scene Generation and Interactive Simulation for Robotic Manipulation

2025-12-01 · Ziqian Wang, Yonghao He, Licheng Yang, Wei Zou 외 arxiv

Simulation provides a low-cost, scalable pathway to large-scale robotic manipulation data collection. However, existing 3D scene generation methods can rarely be applied directly to manipulation data synthesis, as their …

Scene Generation

MesaTask: Towards Task-Driven Tabletop Scene Generation via 3D Spatial Reasoning

2025-09-26 · Jinkun Hao, Naifu Liang, Zhen Luo, Xudong Xu 외 arxiv

The ability of robots to interpret human instructions and execute manipulation tasks necessitates the availability of task-relevant tabletop scenes for training. However, traditional methods for creating these scenes rel…

Spatial ReasoningScene Generation

TableVerse: A Large-scale Tabletop Dataset with Real-world Grounded Layouts for Generalizable Manipulation

2026-07-23 · Boyuan Wang, Yue Zhang, Xutao Xue, Xueyu Song 외 arxiv

The development of generalizable robotic manipulation policies is inherently bounded by the availability of large-scale, high-fidelity scene data. While recent automated synthesis methods attempt to bridge this gap via t…

SAM3D-Phys: Towards Multi-Object Interactive Simulation in Real World

2026-05-28 · Xin Dong, Weijian Deng, Lihan Zhang, Tianru Dai 외 arxiv

This work addresses the problem of recovering complete, simulatable object geometry from reconstructed real-world scenes, enabling physics-based interaction with objects embedded in the scene. While modern multi-view rec…