Data

Tools

Indexes

Applications

Experiments

Open Source Science

deep_residual_voxel_autoencoder

In the domain of computer vision, deep residual neural networks like EfficientNet have set new standards in terms of robustness and accuracy. In this work, we present a deep residual 3D autoencoder based on the EfficientNet architecture for transfer learning. For this purpose, we adopted EfficientNet to 3D problems like voxel models derived from a STEP file.

https://github.com/raoulushpc/deep_residual_voxel_autoencoder

Science Score: 54.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ✓
    CITATION.cff file
    Found CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ○
    DOI references
  • ✓
    Academic publication links
    Links to: arxiv.org
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (6.5%) to scientific vocabulary
Last synced: 10 months ago · JSON representation ·

Repository

In the domain of computer vision, deep residual neural networks like EfficientNet have set new standards in terms of robustness and accuracy. In this work, we present a deep residual 3D autoencoder based on the EfficientNet architecture for transfer learning. For this purpose, we adopted EfficientNet to 3D problems like voxel models derived from a STEP file.

Basic Info
  • Host: GitHub
  • Owner: RaoulusHPC
  • Language: Python
  • Default Branch: main
  • Size: 4.62 MB
Statistics
  • Stars: 8
  • Watchers: 2
  • Forks: 3
  • Open Issues: 1
  • Releases: 0
Created over 4 years ago · Last pushed over 4 years ago
Metadata Files
Readme Citation

README.md

deepresidualvoxel_autoencoder (EfficientNet3D)

Repository for EfficientNets extended to 3D classification and the implementation of an EfficientNet Autoencoder in Tensorflow.

Code follows after paper was accepted to the CIRP Design 2022 Conference

In the domain of computer vision, deep residual neural networks like EfficientNet have set new standards in terms of robustness and accuracy. In this work, we present a deep residual 3D autoencoder based on the EfficientNet architecture for transfer learning. For this purpose, we adopted EfficientNet to 3D problems like voxel models derived from a STEP file.

Paper URL: https://arxiv.org/abs/2202.10099

Cite: R. Schönhof, J. Elstner, R. Manea, S. Tauber, R. Awad, M. F. Huber, Simplified Learning of CAD Features Leveraging a Deep Residual Autoencoder, CIRP Design 2022.

Project Summary:

  • efficientnets: the Efficientnet architecture, currently using the bugfix version
  • lrp: script for testing heatmap generation, not relevant for the repo
  • smalltestscripts: scripts to test ae, not relevant for the repo
  • tensorboards: Saved weights and models from previous training sessions
  • training.py: autonencoder training
  • training_transferlearning: Using the trained encoder, proceed to transferlearning
  • vizualitzation.py: vizualize autoencoder results

Owner

  • Name: Raoul Schönhof
  • Login: RaoulusHPC
  • Kind: user
  • Location: Munich
  • Company: Fraunhofer-Society

Citation (CITATION.cff) 

cff-version: 1.2.0
message: "If you use this software, please cite it as below."
authors:
- family-names: "Schönhof"
  given-names: "Raoul"
  orcid: "https://orcid.org/0000-0002-3003-2324"
- family-names: "Elstner"
  given-names: "Jannes"
- family-names: "Manea"
  given-names: "Radu"
- family-names: "Tauber"
  given-names: "Steffen"
- family-names: "Awad"
  given-names: "Ramez"
- family-names: "Huber"
  given-names: "Marco F."
  
title: "Simplified Learning of CAD Features Leveraging a Deep Residual Autoencoder"
version: 1.0.0
date-released: 2022-02-16
url: "https://github.com/RaoulusHPC/deep_residual_voxel_autoencoder"

preferred-citation:
  type: article
  authors:
  - family-names: "Schönhof"
    given-names: "Raoul"
    orcid: "https://orcid.org/0000-0002-3003-2324"
  - family-names: "Elstner"
    given-names: "Jannes"
  - family-names: "Manea"
    given-names: "Radu"
  - family-names: "Tauber"
    given-names: "Steffen"
  - family-names: "Awad"
    given-names: "Ramez"
  - family-names: "Huber"
    given-names: "Marco F."
  doi: "10.0000/00000"
  journal: "Procedia CIRP Design (Preprint)"
  month: 3
  title: "Simplified Learning of CAD Features Leveraging a Deep Residual Autoencoder"
  year: 2022

GitHub Events

Total
  • Watch event: 1
Last Year
  • Watch event: 1