README.md

metadata

license: mit
pretty_name: Scalable Real2Sim

Scalable Real2Sim

This repository hosts the benchmark dataset and model weights for Scalable Real2Sim: Physics-Aware Asset Generation Via Robotic Pick-and-Place Setups.

• Project website: https://scalable-real2sim.github.io/
• Paper: https://arxiv.org/abs/2503.00370
• Code: https://github.com/nepfaff/scalable-real2sim

Repository Layout

• scalable_real2sim_benchmark_dataset/object_data/*.tar: one archive per object directory under object_data/
• scalable_real2sim_benchmark_dataset/robot_system_id_data/: unpacked robot system identification data
• scalable_real2sim_model_weights/: unpacked model weights used by the pipeline

The object_data subdirectories are stored as individual tar archives to keep the Hugging Face repository at a manageable file count while preserving object-level downloads.

Extraction

To extract all object archives in place:

cd scalable_real2sim_benchmark_dataset/object_data
for archive in *.tar; do
  tar -xf "$archive"
done

After extraction, scalable_real2sim_benchmark_dataset/object_data/ will contain both the original .tar files and the extracted object directories. If you no longer need the archives locally, you can remove them after extraction.

Citation

If you use this dataset or the associated assets, please cite:

@article{pfaff2025_scalable_real2sim,
    author        = {Pfaff, Nicholas and Fu, Evelyn and Binagia, Jeremy and Isola, Phillip and Tedrake, Russ},
    title         = {Scalable Real2Sim: Physics-Aware Asset Generation Via Robotic Pick-and-Place Setups},
    year          = {2025},
    eprint        = {2503.00370},
    archivePrefix = {arXiv},
    primaryClass  = {cs.RO},
    url           = {https://arxiv.org/abs/2503.00370},
}

License

This repository is released under the MIT License. See the accompanying LICENSE file for the full text.