Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://zju3dv.github.io/manhattan_sdf/
Code for "Neural 3D Scene Reconstruction with the Manhattan-world Assumption" CVPR 2022 Oral
https://zju3dv.github.io/manhattan_sdf/
3d-reconstruction 3d-vision computer-vision cvpr2022
Last synced: 3 months ago
JSON representation
Code for "Neural 3D Scene Reconstruction with the Manhattan-world Assumption" CVPR 2022 Oral
- Host: GitHub
- URL: https://zju3dv.github.io/manhattan_sdf/
- Owner: zju3dv
- License: other
- Created: 2022-05-05T18:36:01.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2023-06-06T12:48:58.000Z (over 1 year ago)
- Last Synced: 2024-05-18T19:43:00.476Z (6 months ago)
- Topics: 3d-reconstruction, 3d-vision, computer-vision, cvpr2022
- Language: Python
- Homepage: https://zju3dv.github.io/manhattan_sdf/
- Size: 729 KB
- Stars: 484
- Watchers: 23
- Forks: 34
- Open Issues: 1
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- awesome-planar-reconstruction - [PAGE
README
**News**
* `06/03/2022` We provide the instruction to run on custom data [here](./docs/CUSTOM.md).
* `05/10/2022` To make the comparison on ScanNet easier, we provide all quantitative and qualitative results of baselines [here](./docs/RESULTS.md#baseline-results), including COLMAP, COLMAP*, ACMP, NeRF, UNISURF, NeuS, and VolSDF.
* `05/10/2022` To make the following works easier to compare with our model, we provide our quantitative and qualitative results, as well as the trained models on ScanNet [here](./docs/RESULTS.md#our-results).
* `05/10/2022` We upload our processed ScanNet scene data to [Google Drive](https://drive.google.com/drive/folders/1LyeLuJzTH7-JWUG886gpN1WEzxGDRLm3?usp=sharing).
# Neural 3D Scene Reconstruction with the Manhattan-world Assumption
### [Project Page](https://zju3dv.github.io/manhattan_sdf) | [Video](https://www.youtube.com/watch?v=U4zmSuh31g0) | [Paper](https://arxiv.org/abs/2205.02836)
> [Neural 3D Scene Reconstruction with the Manhattan-world Assumption](https://arxiv.org/abs/2205.02836)
> [Haoyu Guo](https://github.com/ghy0324)\*, [Sida Peng](https://pengsida.net)\*, [Haotong Lin](https://github.com/haotongl), [Qianqian Wang](http://www.cs.cornell.edu/~qqw/), [Guofeng Zhang](http://www.cad.zju.edu.cn/home/gfzhang/), [Hujun Bao](http://www.cad.zju.edu.cn/home/bao/), [Xiaowei Zhou](https://xzhou.me)
> CVPR 2022 (Oral Presentation)
## Setup
### Installation
```shell
conda env create -f environment.yml
conda activate manhattan
```
### Data preparation
Download ScanNet scene data evaluated in the paper from [Google Drive](https://drive.google.com/drive/folders/1LyeLuJzTH7-JWUG886gpN1WEzxGDRLm3?usp=sharing) and extract them into `data/`. Make sure that the path is consistent with [config file](./configs/scannet/0050.yaml#L38).
We provide the instruction to run on custom data [here](./docs/CUSTOM.md).
## Usage
### Training
```shell
python train_net.py --cfg_file configs/scannet/0050.yaml gpus 0, exp_name scannet_0050
```
### Mesh extraction
```shell
python run.py --type mesh_extract --output_mesh result.obj --cfg_file configs/scannet/0050.yaml gpus 0, exp_name scannet_0050
```
### Evaluation
```shell
python run.py --type evaluate --cfg_file configs/scannet/0050.yaml gpus 0, exp_name scannet_0050
```
## Citation
If you find this code useful for your research, please use the following BibTeX entry.
```bibtex
@inproceedings{guo2022manhattan,
title={Neural 3D Scene Reconstruction with the Manhattan-world Assumption},
author={Guo, Haoyu and Peng, Sida and Lin, Haotong and Wang, Qianqian and Zhang, Guofeng and Bao, Hujun and Zhou, Xiaowei},
booktitle={CVPR},
year={2022}
}
```
## Acknowledgement
- Thanks to Lior Yariv for her excellent work [VolSDF](https://lioryariv.github.io/volsdf/).
- Thanks to Jianfei Guo for his implementation of VolSDF [neurecon](https://github.com/ventusff/neurecon).
- Thanks to Johannes Schönberger for his excellent work [COLMAP](https://github.com/colmap/colmap).
- Thanks to Shaohui Liu for his customized implementation of [COLMAP](https://github.com/B1ueber2y/colmap/tree/c84269d693246d8294307cc32f851813f18b6a2d) as a submodule of [NerfingMVS](https://github.com/weiyithu/NerfingMVS).