Thanks to visit codestin.com
Credit goes to github.com

Skip to content

mjmjeong/PhysGaia

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 

Repository files navigation

🌱 PhysGaia: A Physics-Aware Dataset of Multi-Body Interactions for Dynamic Novel View Synthesis

Dataset | Project Page | arXiv Paper

⭐️ Key Highlights

  • 💥 Multi-body interaction
  • 💎 Various materials across all modalities
    • Liquid, Gas, Viscoelastic substance, and Textile
  • ✏️ Physical evaluation
    • physics parameters
    • ground-truth 3D trajectories
  • 😀 Research friendly!!
    • Providing codes for recent Dynamic Novel View Synthesis (DyNVS) models
    • Supporting diverse training setting: both monocular & multiview reconstruction

🔥 Ideal for “Next” Research

  • 🧠 Physical reasoning in dynamic scenes
    • Offering ground-truth physics parameters for precise evaluation of inverse physics estimation
    • Offering ground-truth 3D trajectories for assessing actual motion beyond photorealism.
  • 🤝 Multi-body physical interaction modeling
  • 🧪 Material-specific physics solver integration
  • 🧬 Compatibility with existing DyNVS models

📂 Dataset Structure

Each folder is corresponding to each scene, containing the following files:

{material_type}_{scene_name}.zip
│
├── render/                              # Rendered images
│   ├── train/                           # Images for training
│   └── test/                            # Images for evaluation
│
├── camera_info_test.json                # Monocular camera info for test
├── camera_info_train_mono.json          # Monocular camera info for training
├── camera_info_train_multi.json         # Multi-view camera info for training
│
├── {scene_name}.hipnc                   # Houdini source file (simulation or scene setup)
├── particles/                           # Ground-truth trajectories

For the COLMAP initialization files, please check this drive.

👩🏻‍💻 Code Implementation

Please check each branch for integrated code for recent DyNVS methods.

💳 Citation

TBD

🤝 Contributing

We welcome contributions to expand the dataset (additional modality for new downstream tasks, , implementation for other models, etc.)

Reach out via opening an issue/discussion in the repo.

🛠️ Future Plans

  • Update fidelity of the generated scenes
  • Add more easier scenes: providing more accessible starting points
  • Add guidelines using Houdini source files: ex) How to obtain a flow field?

💳 License

This project is released under the Creative Commons Attribution-NonCommercial 4.0 license.

✅ Free to use, share, and adapt for non-commercial research

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published