MeVGAN: GAN-based plugin model for video generation with applications in colonoscopy.

The generation of videos is crucial, particularly in the medical field, where a significant amount of data is presented in this format. However, due to the extensive memory requirements, creating high-resolution videos poses a substantial challenge for generative models. In this paper, we introduce...

Full description

Saved in:
Bibliographic Details
Main Authors: Łukasz Struski, Tomasz Urbańczyk, Krzysztof Bucki, Bartłomiej Cupiał, Aneta Kaczyńska, Przemysław Spurek, Jacek Tabor
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2025-01-01
Series:PLoS ONE
Online Access:https://doi.org/10.1371/journal.pone.0312038
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849470005653012480
author Łukasz Struski
Tomasz Urbańczyk
Krzysztof Bucki
Bartłomiej Cupiał
Aneta Kaczyńska
Przemysław Spurek
Jacek Tabor
author_facet Łukasz Struski
Tomasz Urbańczyk
Krzysztof Bucki
Bartłomiej Cupiał
Aneta Kaczyńska
Przemysław Spurek
Jacek Tabor
author_sort Łukasz Struski
collection DOAJ
description The generation of videos is crucial, particularly in the medical field, where a significant amount of data is presented in this format. However, due to the extensive memory requirements, creating high-resolution videos poses a substantial challenge for generative models. In this paper, we introduce the Memory Efficient Video GAN (MeVGAN)-a Generative Adversarial Network (GAN) that incorporates a plugin-type architecture. This system utilizes a pre-trained 2D-image GAN, to which we attach a straightforward neural network designed to develop specific trajectories within the noise space. These trajectories, when processed through the GAN, produce realistic videos. We deploy MeVGAN specifically for creating colonoscopy videos, a critical procedure in the medical field, notably helpful for screening and treating colorectal cancer. We show that MeVGAN can produce good quality synthetic colonoscopy videos, which can be potentially used in virtual simulators.
format Article
id doaj-art-46c0d8de329e4bd8becd28c76a23febc
institution Kabale University
issn 1932-6203
language English
publishDate 2025-01-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS ONE
spelling doaj-art-46c0d8de329e4bd8becd28c76a23febc2025-08-20T03:25:16ZengPublic Library of Science (PLoS)PLoS ONE1932-62032025-01-01205e031203810.1371/journal.pone.0312038MeVGAN: GAN-based plugin model for video generation with applications in colonoscopy.Łukasz StruskiTomasz UrbańczykKrzysztof BuckiBartłomiej CupiałAneta KaczyńskaPrzemysław SpurekJacek TaborThe generation of videos is crucial, particularly in the medical field, where a significant amount of data is presented in this format. However, due to the extensive memory requirements, creating high-resolution videos poses a substantial challenge for generative models. In this paper, we introduce the Memory Efficient Video GAN (MeVGAN)-a Generative Adversarial Network (GAN) that incorporates a plugin-type architecture. This system utilizes a pre-trained 2D-image GAN, to which we attach a straightforward neural network designed to develop specific trajectories within the noise space. These trajectories, when processed through the GAN, produce realistic videos. We deploy MeVGAN specifically for creating colonoscopy videos, a critical procedure in the medical field, notably helpful for screening and treating colorectal cancer. We show that MeVGAN can produce good quality synthetic colonoscopy videos, which can be potentially used in virtual simulators.https://doi.org/10.1371/journal.pone.0312038
spellingShingle Łukasz Struski
Tomasz Urbańczyk
Krzysztof Bucki
Bartłomiej Cupiał
Aneta Kaczyńska
Przemysław Spurek
Jacek Tabor
MeVGAN: GAN-based plugin model for video generation with applications in colonoscopy.
PLoS ONE
title MeVGAN: GAN-based plugin model for video generation with applications in colonoscopy.
title_full MeVGAN: GAN-based plugin model for video generation with applications in colonoscopy.
title_fullStr MeVGAN: GAN-based plugin model for video generation with applications in colonoscopy.
title_full_unstemmed MeVGAN: GAN-based plugin model for video generation with applications in colonoscopy.
title_short MeVGAN: GAN-based plugin model for video generation with applications in colonoscopy.
title_sort mevgan gan based plugin model for video generation with applications in colonoscopy
url https://doi.org/10.1371/journal.pone.0312038
work_keys_str_mv AT łukaszstruski mevganganbasedpluginmodelforvideogenerationwithapplicationsincolonoscopy
AT tomaszurbanczyk mevganganbasedpluginmodelforvideogenerationwithapplicationsincolonoscopy
AT krzysztofbucki mevganganbasedpluginmodelforvideogenerationwithapplicationsincolonoscopy
AT bartłomiejcupiał mevganganbasedpluginmodelforvideogenerationwithapplicationsincolonoscopy
AT anetakaczynska mevganganbasedpluginmodelforvideogenerationwithapplicationsincolonoscopy
AT przemysławspurek mevganganbasedpluginmodelforvideogenerationwithapplicationsincolonoscopy
AT jacektabor mevganganbasedpluginmodelforvideogenerationwithapplicationsincolonoscopy