الملخص
Creating deformable assets for robot manipulation requires jointly specifying their geometry, appearance, and physical properties. This is especially challenging for deformable objects, since text and images provide limited evidence about how they deform and respond to contact, yet these responses directly affect their suitability for interaction. Automated generation therefore needs to resolve coupled physical requirements and use interaction evidence to guide construction and refinement. We present DeformSmith, a framework that enables automated generation of interactive, physically credible deformable assets from text or a single image. Through hierarchical agentic construction and a shared physics-grounded harness, it progressively builds, tests, and refines geometry, physical models, material behavior, and robot interaction until the resulting asset is ready for simulation and manipulation. Robot interaction closes the generation loop through manipulation feedback and replayable interaction data. Results show that DeformSmith generates assets with better visual quality and physical plausibility than state-of-the-art baselines, including PhysGen3D, PhysGM, and PhysX-Omni, while supporting the synthesis of data for robotic manipulation of deformable objects. Project page: https://can-lee.github.io/deformsmith-web/
الكلمات المفتاحية
الموضوع
بيانات النشر
- المجلة
- غير متاح
- وصول مفتوح
- وصول مفتوح أخضر
اقتبس هذه المقالة
APA 7
Li, C., Gu, J., Deng, Z., Chen, J., & Sun, L. (2026). DeformSmith: Physics Harness-Guided Hierarchical Generation of Deformable Assets for Robot Manipulation. https://omanscience.com/ar/articles/deformsmith-physics-harness-guided-hierarchical-generation-of-deformable-assets-for-robot-manipulation
MLA 9
Li, Can, et al. "DeformSmith: Physics Harness-Guided Hierarchical Generation of Deformable Assets for Robot Manipulation." https://omanscience.com/ar/articles/deformsmith-physics-harness-guided-hierarchical-generation-of-deformable-assets-for-robot-manipulation.
شيكاغو (المؤلف–التاريخ)
Li, Can, Jie Gu, Zishun Deng, Jingmin Chen, and Lei Sun. 2026. "DeformSmith: Physics Harness-Guided Hierarchical Generation of Deformable Assets for Robot Manipulation." https://omanscience.com/ar/articles/deformsmith-physics-harness-guided-hierarchical-generation-of-deformable-assets-for-robot-manipulation.
هارفارد
Li, C., Gu, J., Deng, Z., Chen, J. and Sun, L. (2026) 'DeformSmith: Physics Harness-Guided Hierarchical Generation of Deformable Assets for Robot Manipulation', Available at: https://omanscience.com/ar/articles/deformsmith-physics-harness-guided-hierarchical-generation-of-deformable-assets-for-robot-manipulation.
فانكوفر
Li C, Gu J, Deng Z, Chen J, Sun L. DeformSmith: Physics Harness-Guided Hierarchical Generation of Deformable Assets for Robot Manipulation. https://omanscience.com/ar/articles/deformsmith-physics-harness-guided-hierarchical-generation-of-deformable-assets-for-robot-manipulation
IEEE
C. Li, J. Gu, Z. Deng, J. Chen, and L. Sun, "DeformSmith: Physics Harness-Guided Hierarchical Generation of Deformable Assets for Robot Manipulation," https://omanscience.com/ar/articles/deformsmith-physics-harness-guided-hierarchical-generation-of-deformable-assets-for-robot-manipulation.