Active View Selection with Perturbed Gaussian Ensemble for Tomographic Reconstruction

Overview

We present Perturbed Gaussian Ensemble, a novel active view selection framework for progressive reconstruction tailored specifically to X-ray Gaussian Splatting. By applying stochastic density perturbations to low-density primitives that are highly susceptible to geometric degradation and measuring the structural disagreement in projection space, our method accurately localizes epistemic uncertainty and predicts the next best view.

Installation

# Download code
git clone https://github.com/yulunwu0108/Perturbed-Gaussian-Ensemble.git --recursive

# Install environment
cd Perturbed-Gaussian-Ensemble
conda env create --file environment.yml
conda activate perturbed-gaussian-ensemble

Dataset

The data used in our experiments will be released soon. Please stay tuned!

Citation

If you find our work useful in your research, please consider citing:

@article{wu2026active,
  title={Active View Selection with Perturbed Gaussian Ensemble for Tomographic Reconstruction},
  author={Wu, Yulun and Zha, Ruyi and Cao, Wei and Li, Yingying and Cai, Yuanhao and Liu, Yaoyao},
  journal={arXiv preprint arXiv:2603.06852},
  year={2026}
}

Acknowledgement

This project has benefited from 3D Gaussian Splatting, R2-Gaussian, FisherRF, and DiffDRR. We sincerely thank the authors for their open-source contributions.