---
_id: '60240'
author:
- first_name: Ingmar
  full_name: Ludwig, Ingmar
  id: '116667'
  last_name: Ludwig
- first_name: Marcel
  full_name: Campen, Marcel
  id: '114904'
  last_name: Campen
  orcid: 0000-0003-2340-3462
citation:
  ama: Ludwig I, Campen M. Strictly Conservative Neural Implicits. <i>Comput Graph
    Forum</i>. 2024;43(7):i–xxii. doi:<a href="https://doi.org/10.1111/CGF.15241">10.1111/CGF.15241</a>
  apa: Ludwig, I., &#38; Campen, M. (2024). Strictly Conservative Neural Implicits.
    <i>Comput. Graph. Forum</i>, <i>43</i>(7), i–xxii. <a href="https://doi.org/10.1111/CGF.15241">https://doi.org/10.1111/CGF.15241</a>
  bibtex: '@article{Ludwig_Campen_2024, title={Strictly Conservative Neural Implicits},
    volume={43}, DOI={<a href="https://doi.org/10.1111/CGF.15241">10.1111/CGF.15241</a>},
    number={7}, journal={Comput. Graph. Forum}, author={Ludwig, Ingmar and Campen,
    Marcel}, year={2024}, pages={i–xxii} }'
  chicago: 'Ludwig, Ingmar, and Marcel Campen. “Strictly Conservative Neural Implicits.”
    <i>Comput. Graph. Forum</i> 43, no. 7 (2024): i–xxii. <a href="https://doi.org/10.1111/CGF.15241">https://doi.org/10.1111/CGF.15241</a>.'
  ieee: 'I. Ludwig and M. Campen, “Strictly Conservative Neural Implicits,” <i>Comput.
    Graph. Forum</i>, vol. 43, no. 7, pp. i–xxii, 2024, doi: <a href="https://doi.org/10.1111/CGF.15241">10.1111/CGF.15241</a>.'
  mla: Ludwig, Ingmar, and Marcel Campen. “Strictly Conservative Neural Implicits.”
    <i>Comput. Graph. Forum</i>, vol. 43, no. 7, 2024, pp. i–xxii, doi:<a href="https://doi.org/10.1111/CGF.15241">10.1111/CGF.15241</a>.
  short: I. Ludwig, M. Campen, Comput. Graph. Forum 43 (2024) i–xxii.
date_created: 2025-06-17T07:46:09Z
date_updated: 2025-06-23T09:01:59Z
department:
- _id: '969'
doi: 10.1111/CGF.15241
extern: '1'
intvolume: '        43'
issue: '7'
language:
- iso: eng
page: i–xxii
publication: Comput. Graph. Forum
status: public
title: Strictly Conservative Neural Implicits
type: journal_article
user_id: '114904'
volume: 43
year: '2024'
...
---
_id: '60333'
abstract:
- lang: eng
  text: <jats:title>Abstract</jats:title><jats:p>We describe HalfedgeCNN, a collection
    of modules to build neural networks that operate on triangle meshes. Taking inspiration
    from the (edge‐based) MeshCNN, convolution, pooling, and unpooling layers are
    consistently defined on the basis of halfedges of the mesh, pairs of oppositely
    oriented virtual instances of each edge. This provides benefits over alternative
    definitions on the basis of vertices, edges, or faces. Additional interface layers
    enable support for feature data associated with such mesh entities in input and
    output as well. Due to being defined natively on mesh entities and their neighborhoods,
    lossy resampling or interpolation techniques (to enable the application of operators
    adopted from image domains) do not need to be employed. The operators have various
    degrees of freedom that can be exploited to adapt to application‐specific needs.</jats:p>
author:
- first_name: Ingmar
  full_name: Ludwig, Ingmar
  id: '116667'
  last_name: Ludwig
- first_name: Daniel
  full_name: Tyson, Daniel
  last_name: Tyson
- first_name: Marcel
  full_name: Campen, Marcel
  id: '114904'
  last_name: Campen
  orcid: 0000-0003-2340-3462
citation:
  ama: Ludwig I, Tyson D, Campen M. HalfedgeCNN for Native and Flexible Deep Learning
    on Triangle Meshes. <i>Computer Graphics Forum</i>. 2023;42(5). doi:<a href="https://doi.org/10.1111/cgf.14898">10.1111/cgf.14898</a>
  apa: Ludwig, I., Tyson, D., &#38; Campen, M. (2023). HalfedgeCNN for Native and
    Flexible Deep Learning on Triangle Meshes. <i>Computer Graphics Forum</i>, <i>42</i>(5).
    <a href="https://doi.org/10.1111/cgf.14898">https://doi.org/10.1111/cgf.14898</a>
  bibtex: '@article{Ludwig_Tyson_Campen_2023, title={HalfedgeCNN for Native and Flexible
    Deep Learning on Triangle Meshes}, volume={42}, DOI={<a href="https://doi.org/10.1111/cgf.14898">10.1111/cgf.14898</a>},
    number={5}, journal={Computer Graphics Forum}, publisher={Wiley}, author={Ludwig,
    Ingmar and Tyson, Daniel and Campen, Marcel}, year={2023} }'
  chicago: Ludwig, Ingmar, Daniel Tyson, and Marcel Campen. “HalfedgeCNN for Native
    and Flexible Deep Learning on Triangle Meshes.” <i>Computer Graphics Forum</i>
    42, no. 5 (2023). <a href="https://doi.org/10.1111/cgf.14898">https://doi.org/10.1111/cgf.14898</a>.
  ieee: 'I. Ludwig, D. Tyson, and M. Campen, “HalfedgeCNN for Native and Flexible
    Deep Learning on Triangle Meshes,” <i>Computer Graphics Forum</i>, vol. 42, no.
    5, 2023, doi: <a href="https://doi.org/10.1111/cgf.14898">10.1111/cgf.14898</a>.'
  mla: Ludwig, Ingmar, et al. “HalfedgeCNN for Native and Flexible Deep Learning on
    Triangle Meshes.” <i>Computer Graphics Forum</i>, vol. 42, no. 5, Wiley, 2023,
    doi:<a href="https://doi.org/10.1111/cgf.14898">10.1111/cgf.14898</a>.
  short: I. Ludwig, D. Tyson, M. Campen, Computer Graphics Forum 42 (2023).
date_created: 2025-06-23T10:34:49Z
date_updated: 2025-07-14T12:48:40Z
department:
- _id: '969'
doi: 10.1111/cgf.14898
extern: '1'
intvolume: '        42'
issue: '5'
language:
- iso: eng
publication: Computer Graphics Forum
publication_identifier:
  issn:
  - 0167-7055
  - 1467-8659
publication_status: published
publisher: Wiley
status: public
title: HalfedgeCNN for Native and Flexible Deep Learning on Triangle Meshes
type: journal_article
user_id: '117512'
volume: 42
year: '2023'
...