Saining Zhang, Baijun Ye, Xiaoxue Chen, Yuantao Chen, Zongzheng Zhang, Cheng Peng, Yongliang Shi, Hao Zhao
We introduce UC-GS, a novel uncertainty-aware 3D-GS training paradigm to effectively use aerial imagery to enhance the NVS of road views.
Our method performs superior on details during viewpoint shifting.
We tested on a server configured with Ubuntu 18.04, cuda 11.6 and gcc 9.4.0. Other similar configurations should also work, but we have not verified each one individually.
- Clone this repo:
git clone https://github.com/SainingZhang/UC-GS.git --recursive
cd UC-GS
- Install dependencies
SET DISTUTILS_USE_SDK=1 # Windows only
conda env create --file environment.yml
conda activate uc_gs
The Synthetic dataset is available in Google Drive.
bash ./single_train.sh
- scene: scene name with a format of
dataset_name/scene_name/orscene_name/; - exp_name: user-defined experiment name;
- gpu: specify the GPU id to run the code. '-1' denotes using the most idle GPU.
- voxel_size: size for voxelizing the SfM points, smaller value denotes finer structure and higher overhead, '0' means using the median of each point's 1-NN distance as the voxel size.
- update_init_factor: initial resolution for growing new anchors. A larger one will start placing new anchor in a coarser resolution.
python render.py -m <path to trained model> # Generate renderings
python metrics.py -m <path to trained model> # Compute error metrics on renderings