/GS-Pull

[NeurIPS'2024] Neural Signed Distance Function Inference through Splatting 3D Gaussians Pulled on Zero-Level Set

Neural Signed Distance Function Inference through Splatting 3D Gaussians Pulled on Zero-Level Set

Wenyuan Zhang · Yu-Shen Liu · Zhizhong Han

NeurIPS 2024

Paper | Project Page

In this paper, we propose to seamlessly combine 3D Gaussians with the learning of neural SDFs. Our method provides a novel perspective to jointly learn 3D Gaussians and neural SDFs by more effectively using multi-view consistency and imposing geometry constraints.

Setup

Training

Code is coming soon.

Evaluation

Pretrained Meshes

Pretrained meshes will be gradually released in Google Drive.

Citation

If you find our code or paper useful, please consider citing

@inproceedings{zhang2024gspull,
    title = {Neural Signed Distance Function Inference through Splatting 3D Gaussians Pulled on Zero-Level Set},
    author = {Wenyuan Zhang and Yu-Shen Liu and Zhizhong Han},
    booktitle = {Advances in Neural Information Processing Systems},
    year = {2024},
}