Home

Awesome

TriVol: Point Cloud Rendering via Triple Volumes

This is the source code of "TriVol: Point Cloud Rendering via Triple Volumes" (CVPR-2023).

<p float="left"> <img src="docs/trivol.png" width="700"/> </p>

Requirements (Tested on 4 * RTX3090)

Installation

Install from environment.yml

You can directly install the requirements through:

$ conda env create -f environment.yml

Or install packages seperately

Training

Training on the ShapeNet Car dataset

Training on the ScanNet dataset

Contact

Tao Hu - taohu@cse.cuhk.edu.hk

Citation

@InProceedings{Hu_2023_CVPR,
    author    = {Hu, Tao and Xu, Xiaogang and Chu, Ruihang and Jia, Jiaya},
    title     = {TriVol: Point Cloud Rendering via Triple Volumes},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month     = {June},
    year      = {2023},
    pages     = {20732-20741}
}