intellabs/baa-ngp

This repository contains the official Implementation for "BAA-NGP: Bundle-Adjusting Accelerated Neural Graphics Primitives".

BAA-NGP: Bundle-Adjusting Accelerated Neural Graphics Primitives

This repository contains the official Implementation for "BAA-NGP: Bundle-Adjusting Accelerated Neural Graphics Primitives".

Installation

Tested on NVIDIA A100 and NVIDIA RTX3090.

Dependencies

• python >= 3.8
• pytorch >= 2.0.1
• tinycudann >= 1.7
• nerfacc >= 0.5.0
• Install the remaining dependencies via pip install -r requirements.txt

Experiments

Blender/nerf_synthetic dataset

• Download nerf_synthetic (1.6G) from here.

  • 100 train images
  • 200 test images

• Train and run

  python baangp/train_baangp.py --scene [lego] --data-root [your_data_root] --save-dir [your_save_dir] --c2f 0.1 0.5
  

Acknowledgements

BAA-NGP code is heavily based on nerfacc and barf.

Citation

If you use this code for your research, please cite our paper BAA-NGP: Bundle-Adjusting Accelerated Neural Graphics Primitives

@article{liu2023baangp,
  title={BAA-NGP: Bundle-Adjusting Accelerated Neural Graphics Primitives.},
  author={Sainan Liu and Shan Lin and Jingpei Lu and Alexey Supikov and Michael Yip},
  journal={CVPRW},
  year={2024}
}

Disclaimer

This “research quality code” is for Non-Commercial purposes and provided by Intel “As Is” without any express or implied warranty of any kind. Please see the dataset's applicable license for terms and conditions. Intel does not own the rights to this data set and does not confer any rights to it. Intel does not warrant or assume responsibility for the accuracy or completeness of any information, text, graphics, links or other items within the code. A thorough security review has not been performed on this code. Additionally, this repository may contain components that are out of date or contain known security vulnerabilities.

nerf_synthetic dataset: Please see the dataset's applicable license for terms and conditions. Intel does not own the rights to this data set and does not confer any rights to it.

Datasets & Models Disclaimer :

To the extent that any public datasets are referenced by Intel or accessed using tools or code on this site those datasets are provided by the third party indicated as the data source. Intel does not create the data, or datasets, and does not warrant their accuracy or quality. By accessing the public dataset(s), or using a model trained on those datasets, you agree to the terms associated with those datasets and that your use complies with the applicable license.

Intel expressly disclaims the accuracy, adequacy, or completeness of any public datasets, and is not liable for any errors, omissions, or defects in the data, or for any reliance on the data. Intel is not liable for any liability or damages relating to your use of public datasets.

baa-ngp's Issues

about no c2f metric

Hi！thanks for your impressive work！I have a question about no c2f metric. When I try this command without --c2f as below. THE metric is better than paper said in lego scene. I want to know if you have any thoughts on this situation？
python baangp/train_baangp.py --scene [lego] --data-root [your_data_root] --save-dir [your_save_dir]

Error reported about CUDA

When I finished my training, CUDA error occurred in the stage of generating training indicators:
RuntimeError: CUDA error: invalid configuration argument
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.
Compile with TORCH_USE_CUDA_DSA to enable device-side assertions.

What is the reason？please!

Regarding the handling and training of the Local Light Field Fusion (LLFF) dataset

Hello sainanl, thank you for your excellent work. I am particularly interested in the part of the code pertaining to the processing and training of the LLFF dataset. May I ask if this section has already been debugged? I would very much appreciate the opportunity to learn about the specific implementation process. Many thanks!