Giter Site home page Giter Site logo

psnet's Introduction

Code for WACV2020 paper "PSNet: A Style Transfer Network for Point Cloud Stylization on Geometry and Color"

Introduction

We perform neural style transer on a point cloud from a point cloud or an image from a style point cloud. The geometry or/and the color property of the content point cloud can be stylized. The color propety can also be stylized from an image. Teaser

Usage

Dependencies

Code was tested on MacOS 10.14 & above and Ubuntu 16.04 with Python3.6.

Install required packages: pip3 install -r requirements.txt.

Be cautious not upgrading matplotlib to 3.1.0 or above, it will drop an error message "It is not currently possible to manually set the aspect on 3D axes" when visualizing point clouds.

Basic usage

Just run main.py. It will style transfer all point clouds in sample_content from each style image or point cloud in sample_style. Results are saved in a new folder style_transfer_results.

Prepare your data

Put your .ply content point clouds in sample_content and your style images or point clouds in sample_style. Then run main.py.

Citation

If you used this code in your publication, please consider citing the following paper:

@InProceedings{Cao_2020_WACV,
author = {Cao, Xu and Wang, Weimin and Nagao, Katashi and Nakamura, Ryosuke},
title = {PSNet: A Style Transfer Network for Point Cloud Stylization on Geometry and Color},
booktitle = {The IEEE Winter Conference on Applications of Computer Vision (WACV)},
month = {March},
year = {2020}
}

Contact

For any questions/comments/bug reports, please feel free to contact [email protected]

psnet's People

Contributors

mfxox avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar  avatar

psnet's Issues

Training code & train using my own data

Hi,

I am very interested in your work. From the code available (i.e. main.py), I found that it loads weights from a trained network model and directly tests on the input point clouds. However, I wish to retrain your network on my own dataset. Is the code for training available in this GitHub repository? If not, will you release it? I am really curious about how it'll look like after being trained on my own dataset.

Regards,
Wei

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.