Neural-Pull: Learning Signed Distance Functions from Point Clouds by Learning to Pull Space onto Surfaces(ICML 2021)

Overview

Neural-Pull: Learning Signed Distance Functions from Point Clouds by Learning to Pull Space onto Surfaces(ICML 2021)

This repository contains the code to reproduce the results from the paper. Neural-Pull: Learning Signed Distance Functions from Point Clouds by Learning to Pull Space onto Surfaces.

You can find detailed usage instructions for training your own models and using pretrained models below.

If you find our code or paper useful, please consider citing

@inproceedings{NeuralPull,
    title = {Neural-Pull: Learning Signed Distance Functions from Point Clouds by Learning to Pull Space onto Surfaces},
    author = {Baorui, Ma and Zhizhong, Han and Yu-shen, Liu and Matthias, Zwicker},
    booktitle = {International Conference on Machine Learning (ICML)},
    year = {2021}
}

Surface Reconstruction Demo

Single Image Reconstruction Demo

Installation

First you have to make sure that you have all dependencies in place. The simplest way to do so, is to use anaconda.

You can create an anaconda environment called tensorflow1 using

conda env create -f NeuralPull.yaml
conda activate tensorflow1

Next, for evaluation of the models,compile the extension modules, which are provided by Occupancy Networks. You can do this via

python setup.py build_ext --inplace

To compile the dmc extension, you have to have a cuda enabled device set up. If you experience any errors, you can simply comment out the dmc_* dependencies in setup.py. You should then also comment out the dmc imports in im2mesh/config.py.

Dataset and pretrained model

  1. You can download our preprocessed data and pretrained model.Included in the link:

    --Our pre-train model on ABC and FAMOUS dataset.

    --Preprocessing data of ABC and FAMOUS(sample points and ground truth points).

    --Our reconstruction results.

  2. To make it easier for you to test the code, we have prepared exmaple data in the exmaple_data folder.

Building the dataset

Alternatively, you can also preprocess the dataset yourself. To this end, you have to follow the following steps:

  • Put your own pointcloud files in 'input_dir' folder, each pointcloud file in a separate .xyz.npy file.
  • Set an empty folder 'out_dir' to place the processed data, note, the folder need to be empty, because this folder will be deleted before the program runs.

You are now ready to build the dataset:

python sample_query_point --out_dir /data1/mabaorui/AtlasNetOwn/data/plane_precompute_2/ --CUDA 0 --dataset other --input_dir ./data/abc_noisefree/04_pts/ 

Training

You can train a new network from scratch, run

  1. Surface Reconstruction
python NeuralPull.py --data_dir /data1/mabaorui/AtlasNetOwn/data/plane_precompute_2/ --out_dir /data1/mabaorui/AtlasNetOwn/plane_cd_sur/ --class_idx 02691156 --train --dataset shapenet
  1. Single Image Reconstruction
python NeuralPull_SVG.py --data_dir /data1/mabaorui/AtlasNetOwn/data/plane_precompute_2/ --out_dir /data1/mabaorui/AtlasNetOwn/plane_cd_sur/ --class_idx 02691156 --train --class_name plane
  1. Train the dataset yourself
python NeuralPull.py --data_dir /data1/mabaorui/AtlasNetOwn/data/plane_precompute_2/ --out_dir /data1/mabaorui/AtlasNetOwn/plane_cd_sur/ --class_idx 02691156 --train --dataset other

Evaluation

For evaluation of the models and generation meshes using a trained model, use

  1. Surface Reconstruction
python NeuralPull.py --data_dir /data1/mabaorui/AtlasNetOwn/data/plane_precompute_2/ --out_dir /data1/mabaorui/AtlasNetOwn/plane_cd_sur/ --class_idx 02691156 --dataset shapenet
  1. Single Image Reconstruction
python NeuralPull_SVG.py --data_dir /data1/mabaorui/AtlasNetOwn/data/plane_precompute_2/ --out_dir /data1/mabaorui/AtlasNetOwn/plane_cd_sur/ --class_idx 02691156 --class_name plane
  1. Evaluation the dataset yourself
python NeuralPull.py --data_dir /data1/mabaorui/AtlasNetOwn/data/plane_precompute_2/ --out_dir /data1/mabaorui/AtlasNetOwn/plane_cd_sur/ --class_idx 02691156 --dataset other

Script Parameters Explanation

Parameters Description
train train or test a network.
data_dir preprocessed data.
out_dir store network parameters when training or to load pretrained network parameters when testing.
class_idx the class to train or test when using shapenet dataset, other dataset, default.
class_name the class to train or test when using shapenet dataset, other dataset, default.
dataset shapenet,famous,ABC or other(your dataset)

Pytorch Implementation of Neural-Pull

Notably, the code in Pytorch implementation is not released by the official lab, it is achieved by @wzxshgz123's diligent work. His intention is only to provide references to researchers who are interested in Pytorch implementation of Neural-Pull. There is no doubt that his unconditional dedication should be appreciated.

A port of muP to JAX/Haiku

MUP for Haiku This is a (very preliminary) port of Yang and Hu et al.'s μP repo to Haiku and JAX. It's not feature complete, and I'm very open to sugg

18 Dec 30, 2022
Generative Flow Networks

Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation Implementation for our paper, submitted to NeurIPS 2021 (also chec

Emmanuel Bengio 381 Jan 04, 2023
A Small and Easy approach to the BraTS2020 dataset (2D Segmentation)

BraTS2020 A Light & Scalable Solution to BraTS2020 | Medical Brain Tumor Segmentation (2D Segmentation) Developed the segmentation models for segregat

Gunjan Haldar 0 Jan 19, 2022
Sub-tomogram-Detection - Deep learning based model for Cyro ET Sub-tomogram-Detection

Deep learning based model for Cyro ET Sub-tomogram-Detection High degree of stru

Siddhant Kumar 2 Feb 04, 2022
PyTorch implementation of "Representing Shape Collections with Alignment-Aware Linear Models" paper.

deep-linear-shapes PyTorch implementation of "Representing Shape Collections with Alignment-Aware Linear Models" paper. If you find this code useful i

Romain Loiseau 27 Sep 24, 2022
Generalized hybrid model for mode-locked laser diodes with an extended passive cavity

GenHybridMLLmodel Generalized hybrid model for mode-locked laser diodes with an extended passive cavity This hybrid simulation strategy combines a tra

Stijn Cuyvers 3 Sep 21, 2022
[ICLR 2021] Is Attention Better Than Matrix Decomposition?

Enjoy-Hamburger 🍔 Official implementation of Hamburger, Is Attention Better Than Matrix Decomposition? (ICLR 2021) Under construction. Introduction T

Gsunshine 271 Dec 29, 2022
Bridging the Gap between Label- and Reference based Synthesis(ICCV 2021)

Bridging the Gap between Label- and Reference based Synthesis(ICCV 2021) Tensorflow implementation of Bridging the Gap between Label- and Reference-ba

huangqiusheng 8 Jul 13, 2022
A library for uncertainty quantification based on PyTorch

Torchuq [logo here] TorchUQ is an extensive library for uncertainty quantification (UQ) based on pytorch. TorchUQ currently supports 10 representation

TorchUQ 96 Dec 12, 2022
The official PyTorch implementation of recent paper - SAINT: Improved Neural Networks for Tabular Data via Row Attention and Contrastive Pre-Training

This repository is the official PyTorch implementation of SAINT. Find the paper on arxiv SAINT: Improved Neural Networks for Tabular Data via Row Atte

Gowthami Somepalli 284 Dec 21, 2022
Tensorflow implementation of Semi-supervised Sequence Learning (https://arxiv.org/abs/1511.01432)

Transfer Learning for Text Classification with Tensorflow Tensorflow implementation of Semi-supervised Sequence Learning(https://arxiv.org/abs/1511.01

DONGJUN LEE 82 Oct 22, 2022
PyTorch code for the ICCV'21 paper: "Always Be Dreaming: A New Approach for Class-Incremental Learning"

Always Be Dreaming: A New Approach for Data-Free Class-Incremental Learning PyTorch code for the ICCV 2021 paper: Always Be Dreaming: A New Approach f

49 Dec 21, 2022
A PyTorch Implementation of SphereFace.

SphereFace A PyTorch Implementation of SphereFace. The code can be trained on CASIA-Webface and the best accuracy on LFW is 99.22%. SphereFace: Deep H

carwin 685 Dec 09, 2022
Official repository of PanoAVQA: Grounded Audio-Visual Question Answering in 360° Videos (ICCV 2021)

Pano-AVQA Official repository of PanoAVQA: Grounded Audio-Visual Question Answering in 360° Videos (ICCV 2021) [Paper] [Poster] [Video] Getting Starte

Heeseung Yun 9 Dec 23, 2022
Code from Daniel Lemire, A Better Alternative to Piecewise Linear Time Series Segmentation

PiecewiseLinearTimeSeriesApproximation code from Daniel Lemire, A Better Alternative to Piecewise Linear Time Series Segmentation, SIAM Data Mining 20

Daniel Lemire 21 Oct 27, 2022
ZeroVL - The official implementation of ZeroVL

This repository contains source code necessary to reproduce the results presente

31 Nov 04, 2022
[CVPR'21] Projecting Your View Attentively: Monocular Road Scene Layout Estimation via Cross-view Transformation

Projecting Your View Attentively: Monocular Road Scene Layout Estimation via Cross-view Transformation Weixiang Yang, Qi Li, Wenxi Liu, Yuanlong Yu, Y

118 Dec 26, 2022
A list of all papers and resoureces on Semantic Segmentation

Semantic-Segmentation A list of all papers and resoureces on Semantic Segmentation. Dataset importance SemanticSegmentation_DL Some implementation of

Alan Tang 1.1k Dec 12, 2022
we propose a novel deep network, named feature aggregation and refinement network (FARNet), for the automatic detection of anatomical landmarks.

Feature Aggregation and Refinement Network for 2D Anatomical Landmark Detection Overview Localization of anatomical landmarks is essential for clinica

aoyueyuan 0 Aug 28, 2022
The source code of CVPR 2019 paper "Deep Exemplar-based Video Colorization".

Deep Exemplar-based Video Colorization (Pytorch Implementation) Paper | Pretrained Model | Youtube video 🔥 | Colab demo Deep Exemplar-based Video Col

Bo Zhang 253 Dec 27, 2022