Meet-in-Style: Text-Driven Real-Time Video Stylization Using Diffusion Models

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F25%3A00383170" target="_blank" >RIV/68407700:21230/25:00383170 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1109/MCG.2025.3554312" target="_blank" >https://doi.org/10.1109/MCG.2025.3554312</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/MCG.2025.3554312" target="_blank" >10.1109/MCG.2025.3554312</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Meet-in-Style: Text-Driven Real-Time Video Stylization Using Diffusion Models

Popis výsledku v původním jazyce

We present Meet-in-Style—a new approach to real-time stylization of live video streams using text prompts. In contrast to previous text-based techniques, our system is able to stylize input video at 30 fps on commodity graphics hardware while preserving structural consistency of the stylized sequence and minimizing temporal flicker. A key idea of our approach is to combine diffusion-based image stylization with a few-shot patch-based training strategy that can produce a custom image-to-image stylization network with real-time inference capabilities. Such a combination not only allows for fast stylization, but also greatly improves consistency of individual stylized frames compared to a scenario where diffusion is applied to each video frame separately. We conducted a number of user experiments in which we found our approach to be particularly useful in video conference scenarios enabling participants to interactively apply different visual styles to themselves (or to each other) to enhance the overall chatting experience.

Název v anglickém jazyce

Meet-in-Style: Text-Driven Real-Time Video Stylization Using Diffusion Models

Popis výsledku anglicky

We present Meet-in-Style—a new approach to real-time stylization of live video streams using text prompts. In contrast to previous text-based techniques, our system is able to stylize input video at 30 fps on commodity graphics hardware while preserving structural consistency of the stylized sequence and minimizing temporal flicker. A key idea of our approach is to combine diffusion-based image stylization with a few-shot patch-based training strategy that can produce a custom image-to-image stylization network with real-time inference capabilities. Such a combination not only allows for fast stylization, but also greatly improves consistency of individual stylized frames compared to a scenario where diffusion is applied to each video frame separately. We conducted a number of user experiments in which we found our approach to be particularly useful in video conference scenarios enabling participants to interactively apply different visual styles to themselves (or to each other) to enhance the overall chatting experience.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/EF16_019%2F0000765" target="_blank" >EF16_019/0000765: Výzkumné centrum informatiky</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2025

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

IEEE Computer Graphics and Applications

ISSN

0272-1716

e-ISSN

1558-1756

Svazek periodika

45

Číslo periodika v rámci svazku

2

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

10

Strana od-do

47-56

Kód UT WoS článku

001508290000017

EID výsledku v databázi Scopus

2-s2.0-105001518462