---
_id: '7002'
abstract:
- lang: eng
text: 'Selecting an appropriate semantic segmentation model for a given application
domain remains a challenging and time-consuming task for practitioners and researchers.
This paper presents an interactive, web-based platform that enables side-by-side
visual comparison of multiple neural network segmentation models applied to identical
images. The system integrates three transformer-based segmentation models: a face-parsing
network producing 19 semantic classes, a SegFormer-B3 clothing segmentation model
with 18 classes, and a Mask2Former model for general-purpose scene segmentation
spanning 150 ADE20K categories. Key contributions include side-by-side evaluation
of model outputs across multiple architectures and image categories, with real-time
segment highlighting and a scalable inference caching system that enables model
comparisons without requiring repeated graphics processing unit (GPU) computation.
The platform organizes a curated dataset of images under a hierarchical category
taxonomy, supporting structured evaluation across demographic and contextual variables.
As a practical use case, the system is applied within the ADRIAN project to assist
in verifying identity consistency across images through segmentation-based analysis.
The platform thus contributes a specialized artificial intelligence (AI) tool
for systematic segmentation and object detection model evaluation within media
analysis pipelines, where selecting appropriate models is a recurring challenge
across tasks from identity verification to content moderation. It is available
under https://github.com/vika-v-v/neural-networks-for-image-segmentation and designed
to lower the barrier for comparative model evaluation in applied computer vision
workflows.'
author:
- first_name: Viktoriia
full_name: Vovchenko, Viktoriia
id: '252442'
last_name: Vovchenko
orcid: 0009-0004-9798-1112
orcid_put_code_url: https://api.orcid.org/v2.0/0009-0004-9798-1112/work/218229773
- first_name: Sergej
full_name: Schultenkämper, Sergej
id: '236164'
last_name: Schultenkämper
orcid: 0009-0005-6858-9813
orcid_put_code_url: https://api.orcid.org/v2.0/0009-0005-6858-9813/work/218229775
- first_name: Frederik
full_name: Bäumer, Frederik
id: '241734'
last_name: Bäumer
orcid: 0000-0002-0826-0144
orcid_put_code_url: https://api.orcid.org/v2.0/0000-0002-0826-0144/work/218229776
citation:
alphadin: 'Vovchenko, Viktoriia ;
Schultenkämper, Sergej ; Bäumer,
Frederik: A Web-Based Platform for Interactive Comparison of Neural Network
Image Segmentation Models. In: Böhm, S.
; Ahrweiler, P. ; Gutenberg,
J. ; Leung, C. (Hrsg.): 2026
The Second International Conference on AI-based Media Innovation : IARIA'
ama: 'Vovchenko V, Schultenkämper S, Bäumer F. A Web-Based Platform for Interactive
Comparison of Neural Network Image Segmentation Models. In: Böhm S, Ahrweiler
P, Gutenberg J, Leung C, eds. 2026 The Second International Conference on AI-Based
Media Innovation. IARIA.'
apa: 'Vovchenko, V., Schultenkämper, S., & Bäumer, F. (n.d.). A Web-Based Platform
for Interactive Comparison of Neural Network Image Segmentation Models. In S.
Böhm, P. Ahrweiler, J. Gutenberg, & C. Leung (Eds.), 2026 The Second International
Conference on AI-based Media Innovation. Nizza, Frankreich: IARIA.'
bibtex: '@inproceedings{Vovchenko_Schultenkämper_Bäumer, title={A Web-Based Platform
for Interactive Comparison of Neural Network Image Segmentation Models}, booktitle={2026
The Second International Conference on AI-based Media Innovation}, publisher={IARIA},
author={Vovchenko, Viktoriia and Schultenkämper, Sergej and Bäumer, Frederik},
editor={Böhm, Stephan and Ahrweiler, Petra and Gutenberg, Johannes and Leung,
Clement Editors} }'
chicago: Vovchenko, Viktoriia, Sergej Schultenkämper, and Frederik Bäumer. “A Web-Based
Platform for Interactive Comparison of Neural Network Image Segmentation Models.”
In 2026 The Second International Conference on AI-Based Media Innovation,
edited by Stephan Böhm, Petra Ahrweiler, Johannes Gutenberg, and Clement Leung.
IARIA, n.d.
ieee: V. Vovchenko, S. Schultenkämper, and F. Bäumer, “A Web-Based Platform for
Interactive Comparison of Neural Network Image Segmentation Models,” in 2026
The Second International Conference on AI-based Media Innovation, Nizza, Frankreich.
mla: Vovchenko, Viktoriia, et al. “A Web-Based Platform for Interactive Comparison
of Neural Network Image Segmentation Models.” 2026 The Second International
Conference on AI-Based Media Innovation, edited by Stephan Böhm et al., IARIA.
short: 'V. Vovchenko, S. Schultenkämper, F. Bäumer, in: S. Böhm, P. Ahrweiler, J.
Gutenberg, C. Leung (Eds.), 2026 The Second International Conference on AI-Based
Media Innovation, IARIA, n.d.'
conference:
end_date: 2026-07-09
location: Nizza, Frankreich
name: 2026 The Second International Conference on AI-based Media Innovation
start_date: 2026-07-05
date_created: 2026-06-19T16:31:47Z
date_updated: 2026-07-27T12:17:01Z
editor:
- first_name: Stephan
full_name: Böhm, Stephan
last_name: Böhm
- first_name: Petra
full_name: Ahrweiler, Petra
last_name: Ahrweiler
- first_name: Johannes
full_name: Gutenberg, Johannes
last_name: Gutenberg
- first_name: 'Clement '
full_name: 'Leung, Clement '
last_name: Leung
keyword:
- segmentation
- model comparison
- face parsing
language:
- iso: eng
main_file_link:
- url: https://www.thinkmind.org/library/AIMEDIA/AIMEDIA_2026
publication: 2026 The Second International Conference on AI-based Media Innovation
publication_identifier:
isbn:
- 978-1-68558-403-0
publication_status: accepted
publisher: IARIA
status: public
title: A Web-Based Platform for Interactive Comparison of Neural Network Image Segmentation
Models
type: conference
user_id: '220548'
year: '2026'
...
---
_id: '6235'
author:
- first_name: Marius
full_name: Sangel, Marius
id: '243822'
last_name: Sangel
- first_name: Emilia
full_name: Bensch, Emilia
last_name: Bensch
- first_name: Hans
full_name: Brandt-Pook, Hans
id: '206531'
last_name: Brandt-Pook
orcid: 0009-0002-6668-2684
orcid_put_code_url: https://api.orcid.org/v2.0/0009-0002-6668-2684/work/194297832
- first_name: Timo
full_name: Röllke, Timo
last_name: Röllke
- first_name: Cedric
full_name: Markworth, Cedric
last_name: Markworth
citation:
alphadin: 'Sangel, Marius ; Bensch, Emilia ; Brandt-Pook,
Hans ; Röllke, Timo ; Markworth, Cedric: Automatisierte Erkennung
von Störstoffen in Bioabfall mit maschinellem Lernen: Ansätze und Ergebnisse aus
dem Projekt TRACES. In: Gesellschaft für
Informatik e.V. (Hrsg.): INFORMATIK 2025. Bonn, 2025, S. 1363–1371'
ama: 'Sangel M, Bensch E, Brandt-Pook H, Röllke T, Markworth C. Automatisierte Erkennung
von Störstoffen in Bioabfall mit maschinellem Lernen: Ansätze und Ergebnisse aus
dem Projekt TRACES. In: Gesellschaft für Informatik e.V., ed. INFORMATIK 2025.
Bonn; 2025:1363-1371. doi:10.18420/INF2025_121'
apa: 'Sangel, M., Bensch, E., Brandt-Pook, H., Röllke, T., & Markworth, C. (2025).
Automatisierte Erkennung von Störstoffen in Bioabfall mit maschinellem Lernen:
Ansätze und Ergebnisse aus dem Projekt TRACES. In Gesellschaft für Informatik
e.V. (Ed.), INFORMATIK 2025 (pp. 1363–1371). Bonn. https://doi.org/10.18420/INF2025_121'
bibtex: '@inproceedings{Sangel_Bensch_Brandt-Pook_Röllke_Markworth_2025, place={Bonn},
title={Automatisierte Erkennung von Störstoffen in Bioabfall mit maschinellem
Lernen: Ansätze und Ergebnisse aus dem Projekt TRACES}, DOI={10.18420/INF2025_121},
number={366}, booktitle={INFORMATIK 2025}, author={Sangel, Marius and Bensch,
Emilia and Brandt-Pook, Hans and Röllke, Timo and Markworth, Cedric}, editor={Gesellschaft
für Informatik e.V.Editor}, year={2025}, pages={1363–1371} }'
chicago: 'Sangel, Marius, Emilia Bensch, Hans Brandt-Pook, Timo Röllke, and Cedric
Markworth. “Automatisierte Erkennung von Störstoffen in Bioabfall mit maschinellem
Lernen: Ansätze und Ergebnisse aus dem Projekt TRACES.” In INFORMATIK 2025,
edited by Gesellschaft für Informatik e.V., 1363–71. Bonn, 2025. https://doi.org/10.18420/INF2025_121.'
ieee: 'M. Sangel, E. Bensch, H. Brandt-Pook, T. Röllke, and C. Markworth, “Automatisierte
Erkennung von Störstoffen in Bioabfall mit maschinellem Lernen: Ansätze und Ergebnisse
aus dem Projekt TRACES,” in INFORMATIK 2025, Potsdam, 2025, no. 366, pp.
1363–1371.'
mla: 'Sangel, Marius, et al. “Automatisierte Erkennung von Störstoffen in Bioabfall
mit maschinellem Lernen: Ansätze und Ergebnisse aus dem Projekt TRACES.” INFORMATIK
2025, edited by Gesellschaft für Informatik e.V., no. 366, 2025, pp. 1363–71,
doi:10.18420/INF2025_121.'
short: 'M. Sangel, E. Bensch, H. Brandt-Pook, T. Röllke, C. Markworth, in: Gesellschaft
für Informatik e.V. (Ed.), INFORMATIK 2025, Bonn, 2025, pp. 1363–1371.'
conference:
end_date: 2025-09-19
location: Potsdam
name: INFORMATIKFESTIVAL 2025
start_date: 2025-09-16
corporate_editor:
- Gesellschaft für Informatik e.V.
date_created: 2025-10-15T07:05:27Z
date_updated: 2026-03-17T15:29:27Z
department:
- _id: 4b2dc5c9-bee3-11eb-b75f-ecc80f94fb21
doi: 10.18420/INF2025_121
file:
- access_level: open_access
content_type: application/pdf
creator: msangel1
date_created: 2025-10-15T07:02:12Z
date_updated: 2025-10-15T07:02:12Z
file_id: '6236'
file_name: Automatisierte Erkennung von Störstoffen in Bioabfall mit maschinellem
Lernen Ansätze und Ergebnisse aus dem Projekt TRACES.pdf
file_size: 14374281
relation: main_file
success: 1
file_date_updated: 2025-10-15T07:02:12Z
has_accepted_license: '1'
issue: '366'
keyword:
- Machine Learning
- Computer Vision
- Instance Segmentation
- CNNs
- YOLACT
- Data Augmentation
- Waste Classification
- Trash Detection
- Biowaste Analysis
language:
- iso: ger
license: https://creativecommons.org/licenses/by-sa/4.0/
main_file_link:
- open_access: '1'
oa: '1'
page: 1363-1371
place: Bonn
project:
- _id: f432a2ee-bceb-11ed-a251-a83585c5074d
name: Institute for Data Science Solutions
publication: INFORMATIK 2025
publication_identifier:
unknown:
- 2944-7682
publication_status: epub_ahead
quality_controlled: '1'
status: public
title: 'Automatisierte Erkennung von Störstoffen in Bioabfall mit maschinellem Lernen:
Ansätze und Ergebnisse aus dem Projekt TRACES'
tmp:
image: /images/cc_by_sa.png
legal_code_url: https://creativecommons.org/licenses/by-sa/4.0/legalcode
name: Creative Commons Attribution-ShareAlike 4.0 International Public License (CC
BY-SA 4.0)
short: CC BY-SA (4.0)
type: conference
urn: urn:nbn:de:hbz:bi10-62353
user_id: '243822'
year: '2025'
...