benchflow/frontierphysics-material-agnostic-visual-inverse-dynamics

573 downloads in 30 days0 likesLast modified 19 Oct 2026View on Hugging Face

The checks could not identify a recognised licence ID in the dataset's metadata.

Declared licence: other, from the tagEvidence: README.md at 6ac72dd
Licence name on the card: mixed-upstream-termsEvidence: README.md at 6ac72dd
Licence files: No licence file at the top level of the repositoryEvidence: file list at 6ac72dd
Declared sources: None listed in the card metadata
Access gated on Hugging Face: No

Each cell links to what was checked: a file, a file list or a Hub record.

Citation text from the card

- **PAC-NeRF data** (`data.zip`, `simulation_data.zip`): Xuan Li, Yi-Ling Qiao, Peter Yichen Chen,
  Krishna Murthy Jatavallabhula, Ming Lin, Chenfanfu Jiang and Chuang Gan,
  [*PAC-NeRF: Physics Augmented Continuum Neural Radiance Fields for Geometry-Agnostic System
  Identification*](https://openreview.net/forum?id=tVkrbkz42vc), ICLR 2023. The
  [PAC-NeRF code](https://github.com/xuan-li/PAC-NeRF) is MIT-licensed. The authors' Box folder
  states no licence for the data, and BenchFlow asserts none. The files are copied unchanged, only
  so that the benchmark can be reproduced. If you are a rights holder and want them removed, open a
  discussion on this repository.
- **Matting checkpoint** (`pytorch_resnet101.pth`): weights from
  [BackgroundMattingV2](https://github.com/PeterL1n/BackgroundMattingV2) (MIT), copied from the
  PAC-NeRF Box folder.
- **MASIV source** (`masiv/…zip`): [yizhou42/MASIV](https://github.com/yizhou42/MASIV), MIT licence
  (`LICENSE.md` inside the archive). Zhao et al., *MASIV: Toward Material-Agnostic System
  Identification from Videos*, arXiv:2508.01112.
- **Pretrained Oracle models** (`oracle/pretrained/`): trained by the task author, Yizhou Zhao, with the MASIV code (MIT) and code derived from [Gaussian Splatting](https://github.com/graphdeco-inria/gaussian-splatting), which is under the Inria and MPII GRAPHDECO licence: research and evaluation use only. `oracle/pretrained/LICENSES.md` has the full notices.

Browse the [FrontierPhysics task datas

All datasets marked Unclear | JSON

The same facts for your agent

A free MCP server at mcp.datalicense.uk/mcp. Connect it from an MCP client such as Claude or Cursor.

dataset_licence({ "id": "benchflow/frontierphysics-material-agnostic-visual-inverse-dynamics" })
signal: "unclear"
stated_licence: "other"
sources: 0