EndoSparse: Real-Time Sparse View Synthesis of Endoscopic Scenes using Gaussian Splatting

Project Page | ArXiv Paper

Accepted by International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2024)

Chenxin Li¹, Brandon Y. Feng^2✉, Yifan Liu¹, Hengyu Liu¹, Cheng Wang¹, Weihao Yu¹, Yixuan Yuan^1✉

¹ The Chinese University of Hong Kong, ² Massachusetts Institute of Technology

^✉ Corresponding Author.

---

💡Highlight

• We present state-of-the-art results on surgical scene reconstruction from a sparse set of endoscopic views, achieving and significantly enhancing the practical usage potential of neural reconstruction methods.
• We demonstrate an effective strategy to instill prior knowledge from a pre-trained 2D generative model to improve and regularize the visual reconstruction quality under sparse observations.
• We introduce an effective strategy to distill geometric prior knowledge from a visual foundation model that drastically improves the geometric reconstruction quality under sparse observations.

🛠Setup

git clone https://github.com/CUHK-AIM-Group/EndoSparse.git
cd EndoSparse
conda create -n endosparse python=3.7
conda activate endosparse

pip install -r requirements.txt

pip install -e submodules/depth-diff-gaussian-rasterization
pip install -e submodules/simple-knn

Tips: 24 GB GPU memory is required for training and inference.

📚Data Preparation

Same to the 📚Data Preparation process of EndoGaussian:

ENDONERF The dataset provided in EndoNeRF is used. You can download and process the dataset from their website. We use the two accessible clips including 'pulling_soft_tissues' and 'cutting_tissues_twice'.

SCARED The dataset provided in SCARED is used. To obtain a link to the data and code release, sign the challenge rules and email them to [email protected]. You will receive a temporary link to download the data and code. Follow MICCAI_challenge_preprocess to extract data. The resulted file structure is as follows.

The file structure is as follows.

├── data
│   | endonerf 
│     ├── pulling
│     ├── cutting 
│   | scared
│     ├── dataset_1
│       ├── keyframe_1
│           ├── data
│       ├── ...
│     ├── dataset_2
|     ├── ...

🎈Acknowledgements

Greatly appreciate the tremendous effort for the following projects!

• FSGS
• EndoGaussian
• 3D GS
• 4D GS
• EndoNeRF

📜Citation

If you find this work helpful for your project,please consider citing the following paper:

@article{li2024endosparse,
  author    = {Chenxin Li and Brandon Y. Feng and Yifan Liu and Hengyu Liu and Cheng Wang and Weihao Yu and Yixuan Yuan},
  title     = {EndoSparse: Real-Time Sparse View Synthesis of Endoscopic Scenes using Gaussian Splatting},
  journal   = {arXiv preprint},
  year      = {2024}
}