Metraipolis › archive › dc_09c9909cdae1

PhysVGGT: Feed-Forward Dense Physical Property Estimation from A Single Image — arXiv:2609.18920

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.18920
License (as archived)
CC0-1.0
Archived
2026-10-15T16:16:44.286Z
SHA-256 of the archived text
9e9c8fa22f1ddb48769135483a7d24690596101ed31c0513a3af10edd9b159fd
Stable URL of this page
https://metraipolis.com/documento/dc_09c9909cdae1

How to cite

PhysVGGT: Feed-Forward Dense Physical Property Estimation from A Single Image — arXiv:2609.18920 — arXiv (en), https://arxiv.org/abs/2609.18920 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_09c9909cdae1 (Metraipolis archive dc_09c9909cdae1, sha256 9e9c8fa22f1ddb48769135483a7d24690596101ed31c0513a3af10edd9b159fd)
BibTeX
@misc{dc09c9909cdae1,
  title = {PhysVGGT: Feed-Forward Dense Physical Property Estimation from A Single Image — arXiv:2609.18920},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.18920},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_09c9909cdae1; sha256 9e9c8fa22f1ddb48769135483a7d24690596101ed31c0513a3af10edd9b159fd},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.18920. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Physical properties, such as friction, hardness, stiffness, and density, govern how robots should grasp, manipulate and interact with objects, yet estimating these properties from RGB images remains challenging. Existing methods typically employ per-object reconstruction augmented with physical properties or directly query vision-language models at test time, which results in substantial computational overhead that limits their applicability. In this work, we present PhysVGGT, a feed-forward model that predicts dense maps of friction coefficient, Shore hardness, Young's modulus, and density, together with object-level mass, from a single RGB image in one forward pass. The key idea of PhysVGGT is to formulate physical property estimation as a dense per-pixel prediction problem and employ a visual geometry transformer to extract geometry-aware tokens from the input image followed by a dense prediction branch for estimating local physical properties and a global prediction branch for estimating object-level mass. In addition, we introduce a scalable pseudo-label generation pipeline that enables large-scale weakly supervised training for dense physical property prediction, substantially reducing the need for expensive direct physical measurements. Extensive experiments show that PhysVGGT achieves state-of-the-art performance on the ABO-500 dataset and generalizes effectively to the out-of-distribution NeRF2Physics dataset. Moreover, PhysVGGT eliminates the need for per-object reconstruction and test-time optimization, achieving an inference latency of only 0.13s per image, making it $27\times$ faster than the previous state of the art.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-17 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2609.18920 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-15T15:45:19.826Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).