Jittor version of "3D Gaussian Splatting"

This repository contains the implementation with jittor for paper "3D Gaussian Splatting for Real-Time Radiance Field Rendering". In our implementation, the speed of evaluation script is 10x faster than PyTorch version.

Set-up

Requirements

jittor
cmake
CUDA>=11
g++
plyfile
tqdm

Compile the submodules

The simple-knn and diff_gaussian_rasterizater modules should be compiled with cmake and make:

cd gaussian-renderer/diff_gaussian_rasterizater
cmake .
make -j
cd ../../scene/simple-knn
cmake .
make -j

You will get simpleknn.so and CudaRasterizer.so in simple-knn and diff_gaussian_rasterizater folders.

LPIPS

The repository uses Jittor_Perceptual-Similarity-Metric for evaluation. Please download the pretrained model following the origin repository and put the weight file in lpips_jittor folder.

Running

To run the optimizer, simply use:

python train.py -s <path to COLMAP or NeRF Synthetic dataset>

Command Line Arguments for train.py

--source_path / -s

Path to the source directory containing a COLMAP or Synthetic NeRF data set.

--model_path / -m

Path where the trained model should be stored (output/<random> by default).

--images / -i

Alternative subdirectory for COLMAP images (images by default).

--eval

Add this flag to use a MipNeRF360-style training/test split for evaluation.

--resolution / -r

Specifies resolution of the loaded images before training. If provided 1, 2, 4 or 8, uses original, 1/2, 1/4 or 1/8 resolution, respectively. For all other values, rescales the width to the given number while maintaining image aspect. If not set and input image width exceeds 1.6K pixels, inputs are automatically rescaled to this target.

--data_device

Specifies where to put the source image data, cuda by default, recommended to use cpu if training on large/high-resolution dataset, will reduce VRAM consumption, but slightly slow down training. Thanks to HrsPythonix.

--white_background / -w

Add this flag to use white background instead of black (default), e.g., for evaluation of NeRF Synthetic dataset.

--sh_degree

Order of spherical harmonics to be used (no larger than 3). 3 by default.

--convert_SHs_python

Flag to make pipeline compute forward and backward of SHs with PyTorch instead of ours.

--convert_cov3D_python

Flag to make pipeline compute forward and backward of the 3D covariance with PyTorch instead of ours.

--debug

Enables debug mode if you experience erros. If the rasterizer fails, a dump file is created that you may forward to us in an issue so we can take a look.

--debug_from

Debugging is slow. You may specify an iteration (starting from 0) after which the above debugging becomes active.

--iterations

Number of total iterations to train for, 30_000 by default.

--ip

IP to start GUI server on, 127.0.0.1 by default.

--port

Port to use for GUI server, 6009 by default.

--test_iterations

Space-separated iterations at which the training script computes L1 and PSNR over test set, 7000 30000 by default.

--save_iterations

Space-separated iterations at which the training script saves the Gaussian model, 7000 30000 <iterations> by default.

--checkpoint_iterations

Space-separated iterations at which to store a checkpoint for continuing later, saved in the model directory.

--start_checkpoint

Path to a saved checkpoint to continue training from.

--quiet

Flag to omit any text written to standard out pipe.

--feature_lr

Spherical harmonics features learning rate, 0.0025 by default.

--opacity_lr

Opacity learning rate, 0.05 by default.

--scaling_lr

Scaling learning rate, 0.005 by default.

--rotation_lr

Rotation learning rate, 0.001 by default.

--position_lr_max_steps

Number of steps (from 0) where position learning rate goes from initial to final. 30_000 by default.

--position_lr_init

Initial 3D position learning rate, 0.00016 by default.

--position_lr_final

Final 3D position learning rate, 0.0000016 by default.

--position_lr_delay_mult

Position learning rate multiplier (cf. Plenoxels), 0.01 by default.

--densify_from_iter

Iteration where densification starts, 500 by default.

--densify_until_iter

Iteration where densification stops, 15_000 by default.

--densify_grad_threshold

Limit that decides if points should be densified based on 2D position gradient, 0.0002 by default.

--densification_interval

How frequently to densify, 100 (every 100 iterations) by default.

--opacity_reset_interval

How frequently to reset opacity, 3_000 by default.

--lambda_dssim

Influence of SSIM on total loss from 0 to 1, 0.2 by default.

--percent_dense

Percentage of scene extent (0--1) a point must exceed to be forcibly densified, 0.01 by default.

Evaluation

By default, the trained models use all available images in the dataset. To train them while withholding a test set for evaluation, use the --eval flag. This way, you can render training/test sets and produce error metrics as follows:

python train.py -s <path to COLMAP or NeRF Synthetic dataset> --eval # Train with train/test split
python render.py -m <path to trained model> # Generate renderings
python metrics.py -m <path to trained model> # Compute error metrics on renderings

Plan of Models

JGaussian will support more valuable 3DGS models in the future, if you are also interested in JGaussian and want to improve it, welcome to submit PR!
✔️Supported 🕒Doing ➕TODO

• ✔️ 3D Gaussian Splatting
• ✔️ Mip-Splatting
• 🕒 FSGS: Real-Time Few-Shot View Synthesis using Gaussian Splatting
• 🕒 2D Gaussian Splatting for Geometrically Accurate Radiance Fields
• ➕ PGSR: Planar-based Gaussian Splatting for Efficient and High-Fidelity Surface Reconstruction
• ➕ EVSplitting: An Efficient and Visually Consistent Splitting Algorithm for 3D Gaussian Splatting
• ➕ ...

Acknowledgements

The original implementation comes from the following cool project:

• 3DGS
• Jittor-LPIPS