https://yilundu.github.io/nerflow/

[ICCV'21] Neural Radiance Flow for 4D View Synthesis and Video Processing
https://yilundu.github.io/nerflow/

Last synced: 8 days ago
JSON representation

[ICCV'21] Neural Radiance Flow for 4D View Synthesis and Video Processing

• Host: GitHub
• URL: https://yilundu.github.io/nerflow/
• Owner: yilundu
• Created: 2021-10-12T03:29:55.000Z (about 3 years ago)
• Default Branch: main
• Last Pushed: 2022-05-01T15:26:51.000Z (over 2 years ago)
• Last Synced: 2024-08-01T04:02:39.406Z (3 months ago)
• Language: Python
• Homepage:
• Size: 3.51 MB
• Stars: 64
• Watchers: 5
• Forks: 18
• Open Issues: 6
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

• awesome-implicit-representations - Neural Radiance Flow for 4D View Synthesis and Video Processing
• awesome-NeRF - Neural Radiance Flow for 4D View Synthesis and Video Processing - and-Beyond.bib#L79-L85) <!---Du20arxiv_nerflow--> (Papers)

README

        
# NeRFlow

[ICCV'21] Neural Radiance Flow for 4D View Synthesis and Video Processing

### Datasets

The pouring dataset used for experiments can be download [at this URL](https://www.dropbox.com/s/bnnjixu1ihyxwn3/pouring_dataset.tar.gz?dl=0) and the iGibson dataset used in experiments can be downloaded [at this URL](https://www.dropbox.com/s/iu12rz0emjp5ija/gibson_dataset.tar?dl=0)

### Pouring Dataset

Please download and extract each dataset at data/nerf\_synthetic/. Please use the following command to train 

```

python run_nerf.py --config=configs/pour_baseline.txt

```

After running model for 200,000 iterations, move the model to a new folder pour\_dataset\_flow and then use the following command

to train with flow consistency

```

python run_nerf.py --config=configs/pour_baseline_flow.txt

```

### Gibson Dataset

Please download and extract each dataset at data/nerf\_synthetic/. Please use the following command to train the model 

```

python run_nerf.py --config=configs/gibson_baseline.txt

```

After running model for 200,000 iterations, move the model to a new folder pour\_dataset\_flow and then use the following command

to train with flow consistency

```

python run_nerf.py --config=configs/gibson_baseline_flow.txt

```

### Citing our Paper

If you find this repo helpful, please consider citing

```	

 @inproceedings{du2021nerflow,

                      author    = {Yilun Du and Yinan Zhang and Hong-Xing Yu 

                                   and Joshua B. Tenenbaum and Jiajun Wu},

                      title     = {Neural Radiance Flow for 4D View Synthesis and Video Processing},

                      year      = {2021},

                      booktitle   = {Proceedings of the IEEE/CVF International Conference

                                     on Computer Vision},

                    }

```