Differential Surfel Rasterization with an additional attribute (part label)

This is the rasterization engine for the paper "Self-supervised Learning of Hybrid Part-aware 3D Representations of 2D Gaussians and Superquadrics". It is modified from the original Differential Surfel Rasterization. Differential Part Surfel Rasterization can support each Gaussian with an additional part attribute, and the part-map can be rendered directly. (Given the time elapsed, I am not sure it works for part attribute optimization via given part-label maps. In our work, we only render part map.)

If you can make use of it in your own research, please be so kind to cite us.

BibTeX

@misc{gao2025selfsupervisedlearninghybridpartaware,
      title={Self-supervised Learning of Hybrid Part-aware 3D Representation of 2D Gaussians and Superquadrics}, 
      author={Zhirui Gao and Renjiao Yi and Yuhang Huang and Wei Chen and Chenyang Zhu and Kai Xu},
      year={2025},
      eprint={2408.10789},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2408.10789}, 
}

how to use

pip install diff-surfel-rasterization_part

use render_part function in render_part.py