Deform360: A Massive Multi-view Visuotactile Dataset for Deformable World Models

Hugging Face Daily Papers Papers

Summary

Deform360 is a large-scale visuotactile dataset with 198 objects and over 215 hours of observations for studying deformable object dynamics, enabling comparison between 2D video and 3D particle world models for robotic manipulation.

Predicting object dynamics (i.e., world modeling) is a fundamental challenge for robotic manipulation, and modeling deformable objects presents a particularly difficult case due to their high-dimensional state spaces and complex material properties. While current world models approach this through two distinct paradigms: learning the dynamics over the 2D pixel space or more explicit 3D geometric space. A systematic understanding of their relative strengths and limitations remains elusive due to the lack of diverse, large-scale real-world data. To address this, we present Deform360, a large-scale visuotactile dataset featuring 198 daily-life objects, 1,980 interaction sequences, and over 215 hours of observations from 41 surround-view cameras and bimanual tactile grippers to capture both global motion and contact-induced local deformations. Leveraging a novel markerless visuotactile 3D tracking pipeline to extract dense geometry and motion, we systematically evaluate current state-of-the-art world models, comparing 2D video models against 3D particle models. Finally, we provide a preliminary demonstration indicating the real-world applicability of our dataset by performing robot planning tasks on deformable objects. Our analysis reveals key insights into the trade-offs between structural priors and scalability, providing a solid benchmark for future research in generalizable deformable object-centric world modeling. Project website: https://deform360.lhy.xyz
Original Article
View Cached Full Text

Cached at: 07/07/26, 06:42 AM

Paper page - Deform360: A Massive Multi-view Visuotactile Dataset for Deformable World Models

Source: https://huggingface.co/papers/2607.05390 Authors:

,

,

,

,

,

,

,

,

,

,

,

,

Abstract

A large-scale visuotactile dataset called Deform360 is introduced to study deformable object dynamics, enabling comparison between 2D video and 3D particle world models for robotic manipulation tasks.

Predicting object dynamics (i.e.,world modeling) is a fundamental challenge for robotic manipulation, and modelingdeformable objectspresents a particularly difficult case due to their high-dimensional state spaces and complex material properties. While current world models approach this through two distinct paradigms: learning the dynamics over the2D pixel spaceor more explicit3D geometric space. A systematic understanding of their relative strengths and limitations remains elusive due to the lack of diverse, large-scale real-world data. To address this, we present Deform360, a large-scalevisuotactile datasetfeaturing 198 daily-life objects, 1,980 interaction sequences, and over 215 hours of observations from 41 surround-view cameras and bimanual tactile grippers to capture both global motion and contact-induced local deformations. Leveraging a novel markerless visuotactile 3D tracking pipeline to extract dense geometry and motion, we systematically evaluate current state-of-the-art world models, comparing 2D video models against3D particle models. Finally, we provide a preliminary demonstration indicating the real-world applicability of our dataset by performingrobot planningtasks ondeformable objects. Our analysis reveals key insights into the trade-offs between structural priors and scalability, providing a solid benchmark for future research in generalizable deformable object-centricworld modeling. Project website: https://deform360.lhy.xyz

View arXiv pageView PDFProject pageGitHub1Add to collection

Get this paper in your agent:

hf papers read 2607\.05390

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2607.05390 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2607.05390 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2607.05390 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

EgoPhys: Learning Generalizable Physics Models of Deformable Objects from Egocentric Video

Hugging Face Daily Papers

EgoPhys introduces a framework to construct deformable physical digital twins from egocentric RGB video using generalizable priors and a compact codebook, enabling zero-shot generalization to unseen objects without per-spring optimization. The system is demonstrated on a real robot, showing that egocentric human play video can serve as internal world representation for deformable-object planning.

τ_0-WM: A Unified Video-Action World Model for Robotic Manipulation

Hugging Face Daily Papers

τ_0-WM is a unified video-action world model for robotic manipulation that integrates policy learning, video prediction, and action evaluation using a shared video diffusion backbone. It shows superior performance on challenging long-horizon and fine-grained tasks.