Coder Social home page Coder Social logo

wx-b / vnn-neural-implicits Goto Github PK

View Code? Open in Web Editor NEW

This project forked from flyinggiraffe/vnn-neural-implicits

0.0 0.0 0.0 2.28 MB

Neural implicit reconstruction experiments for the Vector Neuron paper

License: MIT License

Python 62.58% C++ 14.15% C 8.29% Cuda 12.05% Mako 2.94%
equivariant-network so3 geometric-learning

vnn-neural-implicits's Introduction

Neural Implicit Reconstruction with Vector Neurons

This repository contains code for the neural implicit reconstruction experiments in the paper Vector Neurons: A General Framework for SO(3)-Equivariant Networks. Code for classification and segmentation experiments can be found here.

[Project] [Paper]

Preparation

The code structure follows Occupancy Networks. Please follow their instructions to prepare the data and install the dependencies. Run

python generate_random_rotation.py

to precompute the random rotations for all input pointclouds.

Usage

To train and evaluate the networks, please run these two commands

python train.py CONFIG.yaml
python eval.py CONFIG.yaml

The configuration files are, for VN-OccNet,

configs/equinet/vnn_pointnet_resnet_resnet_ROTATION.yaml

for the vanilla OccNet baseline,

configs/pointcloud/onet_resnet_ROTATION.yaml

and for vanilla PointNet encoder + invariant decoder,

configs/equinet/inner_baseline_resnet_ROTATION.yaml

ROTATION can be chosen from aligned (no rotations) and so3 (with precomputed random rotations). We also provide two settings rot-rand (generate random rotations on the fly during training) and pca (apply PCA pre-alignment the the input pointclouds), which are not reported in the paper.

Citation

Please cite this paper if you want to use it in your work,

@article{deng2021vn,
  title={Vector Neurons: a general framework for SO(3)-equivariant networks},
  author={Deng, Congyue and Litany, Or and Duan, Yueqi and Poulenard, Adrien and Tagliasacchi, Andrea and Guibas, Leonidas},
  journal={arXiv preprint arXiv:2104.12229},
  year={2021}
}

License

MIT License

Acknowledgement

The structure of this codebase is borrowed from Occupancy Networks.

vnn-neural-implicits's People

Contributors

flyinggiraffe avatar

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.