3D Neural Edge Reconstruction

Lei Li · Songyou Peng · Zehao Yu · Shaohui Liu · Rémi Pautrat
Xiaochuan Yin · Marc Pollefeys

CVPR 2024

Paper | Video | Project Page

EMAP enables 3D edge reconstruction from multi-view 2D edge maps.

Installation

git clone https://github.com/cvg/EMAP.git
cd EMAP

conda create -n emap python=3.8
conda activate emap

conda install pytorch==2.1.0 torchvision==0.16.0 torchaudio==2.1.0 pytorch-cuda=12.1 -c pytorch -c nvidia
pip install -r requirements.txt

Datasets

Download datasets:

python scripts/download_data.py 

The data is organized as follows:

<scan_id>
|-- meta_data.json      # camera parameters
|-- color               # images for each view
    |-- 0_colors.png
    |-- 1_colors.png
    ...
|-- edge_DexiNed        # edge maps extracted from DexiNed
    |-- 0_colors.png
    |-- 1_colors.png
    ...
|-- edge_PidiNet        # edge maps extracted from PidiNet
    |-- 0_colors.png
    |-- 1_colors.png
    ...

Training and Edge Extraction

To train and extract edges on different datasets, use the following commands:

ABC-NEF_Edge Dataset

bash scripts/run_ABC.bash

Replica_Edge Dataset

bash scripts/run_Replica.bash

DTU_Edge Dataset

bash scripts/run_DTU.bash

Evaluation

To evaluate extracted edges on ABC-NEF_Edge dataset, use the following commands:

ABC-NEF_Edge Dataset

python src/eval/eval_ABC.py

Code Release Status

• [x] Training Code
• [x] Inference Code
• [x] Evaluation Code
• [ ] Custom Dataset Support

License

Shield:

The majority of EMAP is licensed under a MIT License.

Citing EMAP

If you find the code useful, please consider the following BibTeX entry.

@InProceedings{li2024neural,
  title={3D Neural Edge Reconstruction},
  author={Li, Lei and Peng, Songyou and Yu, Zehao and Liu, Shaohui and Pautrat, R{\'e}mi and Yin, Xiaochuan and Pollefeys, Marc},
  booktitle={IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2024},
}

Contact

If you encounter any issues, you can also contact Lei through [email protected].

Acknowledgement

This project is built upon NeuralUDF, NeuS and MeshUDF. We use pretrained DexiNed and PidiNet for edge map extraction. We thank all the authors for their great work and repos.