CVPR 2026 Findings

Project Page | Paper

Tianhao Xie^{1,*, †}, Linlian Jiang^1,2,*, Xinxin Zuo¹, Yang Wang^1,2, Tiberiu Popa¹

¹Concordia University, Montreal, ²Mila-Quebec AI Institute

^*Equal Contribution ^†Corresponding Author

To create a conda environment with the same dependencies as the codebase, please run:

conda env create -f environment.yml 

Then, clone this repository and install the codebase by running:

git clone https://github.com/tianhaoxie/FACT-GS.git
cd FACT-GS
python -m pip install -e . # install in editable mode for development

The examples of training on NeRF synthetic dataset and Tanks & Temples dataset were provided in ./examples/run_exps.py

This is the official implementation of the paper FACT-GS: Frequency-Aligned Complexity-Aware Texture Reparameterization for 2D Gaussian Splatting. This project was built upon Textured GS with adding the custom cuda kernels for FACT-GS.