VideoRFSplat (ICCV 2025)

This repository contains the official pytorch implementation of the paper: "VideoRFSplat: Direct Scene-Level Text-to-3D Gaussian Splatting Generation with Flexible Pose and Multi-View Joint Modeling".

Code will be released soon.

TODO:

• Inference code
  • Memory efficient implementation
• Text annotation release (synced with https://github.com/gohyojun15/SplatFlow)
• Training script

Citation

If you find this repository helpful for your project, please consider citing our work. :)

@article{go2025videorfsplat,
  title={VideoRFSplat: Direct Scene-Level Text-to-3D Gaussian Splatting Generation with Flexible Pose and Multi-View Joint Modeling},
  author={Go, Hyojun and Park, Byeongjun and Nam, Hyelin and Kim, Byung-Hoon and Chung, Hyungjin and Kim, Changick},
  journal={arXiv preprint arXiv:2503.15855},
  year={2025}
}