3D Gaussian Splatting for Real-Time Radiance Field Rendering (using results of PlanarSLAM)

This repo is cloned from gaussian-splatting. We add additional functions to read Point Clouds, Camera Poses, and RGB Images from results of PlanarSLAM. We hope more possibilities can be explored based on the new Point Clouds.

Motivation: Why use results of PlanarSLAM

New Features of this type of input

1. Points lying on the non-textured regions
2. Global plane instances that are represented in different colors
3. Surface normal vector of every point

BibTex

@Article{kerbl3Dgaussians,
      author       = {Kerbl, Bernhard and Kopanas, Georgios and Leimk{\"u}hler, Thomas and Drettakis, George},
      title        = {3D Gaussian Splatting for Real-Time Radiance Field Rendering},
      journal      = {ACM Transactions on Graphics},
      number       = {4},
      volume       = {42},
      month        = {July},
      year         = {2023},
      url          = {https://repo-sam.inria.fr/fungraph/3d-gaussian-splatting/}
}

@inproceedings{Li2021PlanarSLAM,
  author = {Li, Yanyan and Yunus, Raza and Brasch, Nikolas and Navab, Nassir and Tombari, Federico},
  title = {RGB-D SLAM with Structural Regularities},
  year = {2021},
  booktitle = {2021 IEEE international conference on Robotics and automation (ICRA)},
 }

Setup

We follow the same install steps as suggested by the Original README document.