Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://zju3dv.github.io/im4d/

SIGGRAPH Asia 2023: Code for "Im4D: High-Fidelity and Real-Time Novel View Synthesis for Dynamic Scenes"
https://zju3dv.github.io/im4d/

4d-reconstruction dynamic-view-synthesis siggraph-asia-2023 view-synthesis

Last synced: 3 months ago
JSON representation

SIGGRAPH Asia 2023: Code for "Im4D: High-Fidelity and Real-Time Novel View Synthesis for Dynamic Scenes"

Awesome Lists containing this project

README 

        
# Im4D: High-Fidelity and Real-Time Novel View Synthesis for Dynamic Scenes



### [Project Page](https://zju3dv.github.io/im4d) | [Paper](https://arxiv.org/abs/2310.08585) | [Video](https://youtu.be/pPl1M5jpK4g)

> [High-Fidelity and Real-Time Novel View Synthesis for Dynamic Scenes](https://arxiv.org/abs/2310.08585) \

> Haotong Lin, Sida Peng, Zhen Xu, Tao Xie, Xingyi He, Hujun Bao and Xiaowei Zhou \

> SIGGRAPH Asia 2023 conference track



![DNA-Rendering](https://github.com/haotongl/imgbed/raw/master/im4d/renbody.gif)



## Installation



### Set up the python environment

 Tested with an Ubuntu workstation i9-12900K, 3090GPU



```

conda create -n im4d python=3.10

conda activate im4d

conda install pytorch torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia # pytorch 2.0.1

pip install git+https://github.com/NVlabs/tiny-cuda-nn/#subdirectory=bindings/torch 

pip install -r requirments.txt

```



### Set up datasets



 0. Set up workspace



The workspace is the disk directory that stores datasets, training logs, checkpoints and results. Please ensure it has enough disk space. 



```

export workspace=$PATH_TO_YOUR_WORKSPACE

```



 1. Prepare ZJU-MoCap and NHR datasets.



Please refer to [mlp_maps](https://github.com/zju3dv/mlp_maps/blob/master/INSTALL.md) to download ZJU-MoCap and NHR datasets.

After downloading, place them into `$workspace/zju-mocap` and `$workspace/NHR`, respectively.



 2. Prepare the DNA-Rendering dataset.



Since the license of the [DNA-Rendering](https://dna-rendering.github.io/index.html) dataset does not allow distribution, we cannot release the processed dataset publicly.

You can download the DNA-Rendering dataset at [here](https://dna-rendering.github.io/inner-download.html) or [OpenXLab](https://openxlab.org.cn/datasets/OpenXDLab/DNA-Rendering) . If someone is interested at the processed data, please email me ([email protected]).

You need to cite [DNA-Rendering](https://dna-rendering.github.io/index.html) if you find this data useful.

 



### Pre-trained models



Download pre-trained models from [this link](https://drive.google.com/drive/folders/1_huSP1XOG-HttZwu-JxmICrsR9YQOpkm?usp=sharing) for quick test. Place FILENAME.pth into\

`$workspace/trained_model/SCENE/im4d/FILENAME/latest.pth`. \

e.g., my_313.pth -> `$workspace/trained_model/my_313/im4d/my_313/latest.pth` \

my_313_demo.pth -> `$workspace/trained_model/my_313/im4d/my_313_demo/latest.pth`.



## Testing



 1. Reproduce the quantitative results in the paper.



```

python run.py --type evaluate --cfg_file configs/exps/im4d/xx_dataset/xx_scene.yaml save_result True

```



For the NHR dataset, please firstly download [the preprocessed data](https://drive.google.com/drive/folders/1rA1gzzub6TkGIuu-LaqYwwwiJm4svK2F?usp=sharing) and place them into `$workspace/evaluation`. This evaluation setting is taken from [mlp_maps](https://zju3dv.github.io/mlp_maps/).

Then run one more command to report the PSNR metric:

```

python scripts/evaluate/im4d/eval_nhr.py --gt_path $workspace/evaluation/sport_1_easymocap --output_path $workspace/result/sport_1_easymocap/im4d/sport1_release/default/step00999999/rgb_0

```



 2. Accelerate the rendering speed .

First, precompute the binary fields.



```

python run.py --type cache_grid --cfg_file configs/exps/im4d/renbody/0013_01.yaml --configs configs/components/opts/cache_grid.yaml grid_tag default

```

You may need to change the frames and grid_resolution to fit your scene. 

For example, the scene in ZJU-MoCap has 300 frames and its height is z-axis:

```

python run.py --type cache_grid --cfg_file configs/exps/im4d/zju/my_313.yaml --configs configs/components/opts/cache_grid.yaml grid_tag default grid_resolution 128,128,256 test_dataset.frame_sample 0,300,1

```



Then, render images with the precomputed binary fields.



```

python run.py --type evaluate --cfg_file configs/exps/im4d/renbody/0013_01.yaml --configs configs/components/opts/fast_render.yaml grid_tag default save_result True

```



You may try slightly decreasing sigma_thresh (default: 5.0) to preserve more voxels.



 3. Render a video with the selected trajectory.



```

python run.py --type evaluate --cfg_file configs/exps/im4d/renbody/0013_01.yaml --configs configs/components/opts/render_path/renbody_path.yaml

```

We can render it with the precomputed binary fields by adding one more argument:



```

python run.py --type evaluate --cfg_file configs/exps/im4d/renbody/0013_01.yaml --configs configs/components/opts/render_path/renbody_path.yaml --configs configs/components/opts/fast_render.yaml

```



For better performance, you can use our pre-trained demo models which are trained with all camera views.



```

python run.py --type evaluate --cfg_file configs/exps/im4d/zju/my_313.yaml   --configs configs/components/opts/fast_render.yaml --configs configs/components/opts/render_path/zju_path.yaml exp_name_tag demo

```



## Training



```

python train_net.py --cfg_file configs/exps/im4d/xx_dataset/xx_scene.yaml

```



Training with multiple GPUs:

```

export CUDA_VISIBLE_DEVICES=0,1,2,3

export NUM_GPUS=4

export LOG_LEVEL=WARNING # INFO, DEBUG, WARNING

torchrun --nproc_per_node=$NUM_GPUS train_net.py --cfg_file configs/exps/im4d/xx_dataset/xx_scene.yaml --log_level $LOG_LEVEL distributed True

```



## Running on the custom dataset



For both studio dataset and wild dataset, you can refer to [this tutorial](./docs/custom_data_tutorial.md).



## Acknowledgements

We would like to acknowledge the following inspring prior work:

- [IBRNet: Learning Multi-View Image-Based Rendering](https://ibrnet.github.io/) (Wang et al.)

- [ENeRF: Efficient Neural Radiance Fields for Interactive Free-viewpoint Video](https://zju3dv.github.io/enerf) (Lin et al.)

- [K-Planes: Explicit Radiance Fields in Space, Time, and Appearance](https://sarafridov.github.io/K-Planes/) (Fridovich-Keil et al.)



Big thanks to [NeRFAcc](https://www.nerfacc.com/) (Li et al.) for their efficient implementation, which has significantly accelerated our rendering.



Recently, in the course of refining our codebase, we have incorporated basic implementations of ENeRF and K-Planes. These additions, although not yet thoroughly tested and aligned with the official codes, could serve as useful resources for further exploration and development.

## Citation



If you find this code useful for your research, please use the following BibTeX entry



```

@inproceedings{lin2023im4d,

  title={High-Fidelity and Real-Time Novel View Synthesis for Dynamic Scenes},

  author={Lin, Haotong and Peng, Sida and Xu, Zhen and Xie, Tao and He, Xingyi and Bao, Hujun and Zhou, Xiaowei},

  booktitle={SIGGRAPH Asia Conference Proceedings},

  year={2023}

}

```